new world hydrological cycle observing system (whycos) - … · 2009. 11. 13. · 7.3...

WORLD METEOROLOGICAL ORGANIZATION

TECHNICAL NOTE No. 201

AGROMETEOROLOGYRELATED TO EXTREME EVENTS

by

H.P. Das, T.I. Adamenko, K.A. Anaman, R.G. Gommes and G. Johnson

(CAgM-XI Working Group on Agrometeorology Related to Extreme Events)

Secretariat of the World Meteorological Organization - Geneva - Switzerland

WMO-No. 943

SERNA_B

Copyright in this electronic file and its contents is vested in WMO. It must not be altered, copied or passed on to a third party or posted electronically without WMO's written permission.

The World Meteorological Organization

The World Meteorological Organization (WMO), of which 185* States and Territories are Members, is a specialized agencyof the United Nations. The purposes of the Organization are:

(a) To facilitate world-wide cooperation in the establishment of networks of stations for the making of meteorologicalobservations as well as hydrological and other geophysical observations related to meteorology, and to promote theestablishment and maintenance of centres charged with the provision of meteorological and related services;

(b) To promote the establishment and maintenance of systems for the rapid exchange of meteorological and relatedinformation;

(c) To promote standardization of meteorological and related observations and to ensure the uniform publication ofobservations and statistics;

(d) To further the application of meteorology to aviation, shipping, water problems, agriculture and other human activities;(e) To promote activities in operational hydrology and to further close cooperation between Meteorological and Hydrological

Services; and(f) To encourage research and training in meteorology and, as appropriate, in related fields and to assist in

coordinating the international aspects of such research and training.

(Convention of the World Meteorological Organization, Article 2)

The Organization consists of the following:

The World Meteorological Congress, the supreme body of the Organization, brings together the delegates of Members onceevery four years to determine general policies for the fulfilment of the purposes of the Organization, to approve long-termplans, to authorize maximum expenditures for the following financial period, to adopt Technical Regulations relating tointernational meteorological and operational hydrological practice, to elect the President and Vice-Presidents of theOrganization and members of the Executive Council and to appoint the Secretary-General;

The Executive Council, composed of 36 directors of national Meteorological or Hydrometeorological Services, meets at leastonce a year to review the activities of the Organization and to implement the programmes approved by Congress;

The six regional associations (Africa, Asia, South America, North and Central America, South-West Pacific and Europe),composed of Members, coordinate meteorological and related activities within their respective Regions;

The eight technical commissions, composed of experts designated by Members, study matters within their specific areas ofcompetence (technical commissions have been established for basic systems, instruments and methods of observation,atmospheric sciences, aeronautical meteorology, agricultural meteorology, marine meteorology, hydrology, and climatology);

The Secretariat, headed by the Secretary-General, serves as the administrative, documentation and information centre of theOrganization. It prepares, edits, produces and distributes the publications of the Organization, carries out the duties specifiedin the Convention and other Basic Documents and provides secretariat support to the work of the constituent bodies of WMOdescribed above.

________* On 1 July 2002

WORLD METEOROLOGICAL ORGANIZATION

Secretariat of the World Meteorological Organization – Geneva – Switzerland2003

WMO-No. 943

TECHNICAL NOTE No. 201

AGROMETEOROLOGYRELATED TO EXTREME EVENTS

by

H.P. Das, T.I. Adamenko, K.A. Anaman, R.G. Gommes and G. Johnson

(CAgM-XI Working Group on Agrometeorology Related to Extreme Events)

© 2003, World Meteorological Organization

ISBN: 92-63-10943-5

NOTE

The designations employed and the presentation of material in this publication do not implythe expression of any opinion whatsoever on the part of the Secretariat of the World Meteorological Organization concerning the legal status of any country, territory, city or area, or of its authorities, or concerning the delimitation of its frontiers or boundaries.

CONTENTS

Page

FOREWORD ............................................................................................................................. VII

SUMMARY (English, French, Russian and Spanish)............................................................... IX

LIST OF CONTRIBUTORS .................................................................................................... XIII

CHAPTER 1 — INTRODUCTION (by H.P. Das) ................................................................. 11.1 Definition of extreme events ......................................................................................... 11.2 Types of extreme climatic events ................................................................................... 11.3 Agrometeorological data related to extreme events ...................................................... 11.4 Extreme events and agricultural production ................................................................. 31.5 Socio-economic impact of extreme events .................................................................... 41.6 Prevention and preparedness ......................................................................................... 51.7 Rehabilitation ................................................................................................................ 5References .................................................................................................................................. 6

CHAPTER 2 — AGROMETEOROLOGICAL ASPECTS OF DROUGHT ANDDESERTIFICATION (by H.P. Das) .......................................................................................... 72.1 Drought ........................................................................................................................... 7

2.1.1 Introduction ................................................................................................... 72.1.2 Drought and famine ....................................................................................... 72.1.3 Drought concepts, definitions and quantifications........................................ 82.1.4 Data availability ............................................................................................. 92.1.5 Causes of drought ........................................................................................... 102.1.6 Spatial and temporal aspects of drought ........................................................ 112.1.7 Impact of drought ........................................................................................... 112.1.8 Forecasting drought ........................................................................................ 132.1.9 Drought detection, monitoring and early warning ........................................ 132.1.10 Adaptation and adjustments to drought ........................................................ 162.1.11 Drought management: mitigation, preparedness and policy ......................... 172.1.12 Summary and conclusion ............................................................................... 19

2.2 Desertification................................................................................................................. 192.2.1 Introduction ................................................................................................... 192.2.2 Definition of desertification ........................................................................... 202.2.3 Distribution of desertification ........................................................................ 212.2.4 Desertification trends ..................................................................................... 212.2.5 Physical processes of desertification ............................................................... 232.2.6 Causes of desertification................................................................................. 232.2.7 Desertification and feedback mechanism....................................................... 262.2.8 Desertification and development ................................................................... 272.2.9 Monitoring and assessment of desertification ................................................ 272.2.10 Recovery and control of desertification ......................................................... 282.2.11 Summary and conclusion ............................................................................... 292.2.12 Suggestions and recommendations ................................................................ 30

References .................................................................................................................................. 30

CHAPTER 3 — INCIDENCE, PREDICTION, MONITORING AND MITIGATIONMEASURES OF TROPICAL CYCLONES AND STORM SURGES (by H.P. Das)............. 353.1 Introduction .................................................................................................................... 353.2 Geographical distribution of tropical cyclones .............................................................. 353.3 Regional categorization of tropical cyclones and their intensity ................................... 373.4 Characteristics of a tropical cyclone............................................................................... 373.5 Conditions necessary for tropical cyclone formation ..................................................... 37

3.6 Storm surge ..................................................................................................................... 383.6.1 Protection from the storm surge..................................................................... 38

3.7 Heavy rains associated with a hurricane/typhoon.......................................................... 383.8 Surface wind in a tropical cyclone.................................................................................. 393.9 Availability of data for monitoring and forecasting tropical cyclone............................. 403.10 Destruction caused by tropical storms ............................................................................ 40

3.10.1 Damage to agriculture .................................................................................... 403.10.2 Salt deposition in coastal areas ...................................................................... 403.10.3 Agrometeorological loss associated with some devastating cyclones ............ 403.10.4 Other destructive effects of cyclones.............................................................. 433.10.5 Some economic and social consequences ...................................................... 43

3.11 Beneficial impacts of cyclonic storms ............................................................................. 443.12 Cyclone warning system ................................................................................................. 44

3.12.1 Dissemination of cyclone warning ................................................................. 453.13 Disaster management and mitigation measures.............................................................. 45

3.13.1 Prevention ...................................................................................................... 453.13.2 Preparedness ................................................................................................... 463.13.3 Evacuation...................................................................................................... 463.13.4 Mitigation....................................................................................................... 47

3.14 Other measures ............................................................................................................... 483.15 Summary and conclusion................................................................................................ 483.16 Recommendations .......................................................................................................... 49References .................................................................................................................................. 49

CHAPTER 4 — ASSESSING THE ECONOMIC AND SOCIAL IMPACTS OFEXTREME EVENTS ON AGRICULTURE AND THE USE OF METEOROLOGICALINFORMATION TO REDUCE ADVERSE IMPACTS (by K.A. Anaman).......................... 524.1 Social and economic impacts of extreme events affecting agriculture........................... 52

4.1.1 Classification of extreme events affecting agriculture and rural society........ 524.1.2 Assessment of economic and social impacts of extreme events .................... 52

4.2 Economic use of meteorological information and services to reduce adverseimpacts of extreme events on agriculture ....................................................................... 554.2.1 Data, information and services as economic resources .................................. 554.2.2 Desirable attributes of meteorological information and services ................... 584.2.3 Introduction to the economic theory of markets........................................... 604.2.4 Valuation of meteorological information and services based on benefits

accruing to producers and consumers of commodities that utilizemeteorological information as inputs in their production processes ............. 63

4.2.5 Economic valuation of meteorological data as environmental resources...... 66References .................................................................................................................................. 69

CHAPTER 5 — ASSESSING THE IMPACT OF EXTREME WEATHER ANDCLIMATE EVENTS ON AGRICULTURE, WITH PARTICULAR REFERENCE TOFLOODING AND HEAVY RAINFALL (by G. Johnson)....................................................... 735.1 Survey of countries’ assessments of extreme weather and climate impacts,

focusing on flooding and heavy rainfall.......................................................................... 735.1.1 Overview of survey and analysis..................................................................... 73

5.2 The impact of flooding and heavy rainfall on agriculture.............................................. 815.2.1 Overview ........................................................................................................ 815.2.2 Characteristics of flooding and/or heavy rainfall as extreme

agrometeorological events ............................................................................. 825.2.3 Geographic and topographic considerations.................................................. 865.2.4 Data and analyses for assessments, planning and mitigation......................... 875.2.5 Examples of flood and heavy rainfall impacts on agriculture focusing on

the USA. Assessments, prediction, warning, monitoring and mitigation..... 88References .................................................................................................................................. 99Appendix:Table A.1 Questionnaire on agrometeorology related to extreme events ............... 101

IV CONTENTS

Page

CHAPTER 6 — HAIL, HIGH WINDS AND COLD INJURY (by T.I. Adamenko)............. 1076.1 Hail.................................................................................................................................. 107

6.1.1 Measures to protect against hailstorms .......................................................... 1106.2 High winds ...................................................................................................................... 110

6.2.1 Dust storms ..................................................................................................... 1116.2.2 Counteracting dust storms.............................................................................. 114

6.3 Extreme cold weather including cold injury................................................................... 114References .................................................................................................................................. 118

CHAPTER 7 — LOCUSTS (by R.G. Gommes)...................................................................... 1197.1 Introduction .................................................................................................................... 1197.2 Development of locusts................................................................................................... 1197.3 Locust-climate interactions ............................................................................................ 1197.4 Some locusts.................................................................................................................... 120

7.4.1 Desert locusts.................................................................................................. 1207.4.2 Australian plague locusts................................................................................ 1207.4.3 Migratory locust.............................................................................................. 121

7.5 Monitoring of acridian situations ................................................................................... 121References .................................................................................................................................. 122

CHAPTER 8 — SPECIFICATION FOR A DATABASE OF EXTREMEAGROMETEOROLOGICAL EVENTS (by R.G. Gommes) .................................................. 1238.1 Introduction .................................................................................................................... 1238.2 Categories covered .......................................................................................................... 124

8.2.1 Direct natural atmospheric factors ................................................................. 1248.2.2 Indirect natural atmospheric factors and complex interactions .................... 1248.2.3 Man-made factors ........................................................................................... 1258.2.4 Other geophysical factors ............................................................................... 1268.2.5 Other non-geophysical factors ....................................................................... 1298.2.6 Very rare factors.............................................................................................. 129

8.3 Information to be stored in the database........................................................................ 1308.3.1 Definition of an event .................................................................................... 1308.3.2 The three database components – thematic description of impacted system ..... 1318.3.3 Sources of data................................................................................................ 132

8.4 Conclusions and recommendations................................................................................ 132References .................................................................................................................................. 133

CHAPTER 9 — CONCLUSIONS AND RECOMMENDATIONS (by H.P. Das)................ 1359.1 Conclusions..................................................................................................................... 1359.2 Recommendations .......................................................................................................... 136

9.2.1 Information systems........................................................................................ 1369.2.2 Monitoring, early warning and remedial measures ........................................ 1369.2.3 Methodology development ............................................................................ 1379.2.4 Training, education and increased awareness for the general public

and decision makers ....................................................................................... 1379.2.5 Collaboration and cooperation ...................................................................... 137

CONTENTS V

Page

Extreme events affect human society and cause suffering, damage to infrastructure worth billions of dollars each year, lossof life and deterioration of the environment, substantially exceeding normal expectations. In recent decades, peoplethroughout the world have become increasingly concerned by extreme meteorological and hydrological events, which arebecoming more frequent and more destructive. The extreme events that affect agriculture are mainly those that are relatedto weather and climate such as drought, floods, extreme hot and dry weather, frost, excessive rainfall, tropical cyclones,storm surges, high winds, hailstorms, heat stress and cold injuries.

Indeed few communities are immune to these events although some communities are more vulnerable to particularevents than others. In the developing world, some of the impacts that can accompany extreme meteorological andhydrological events include the decline in agricultural production and destruction of food reserves and damage to or lossof water supplies through drought or through pollution of traditional water sources during floods. The extension ofcultivation into less suitable climates also increases the risk of damage.

Accurate information on extreme meteorological and hydrological events can therefore help farmers takepreventive steps to limit damage and increase agricultural output. The information is also useful for several other purposesincluding modification of the crop environment, protection from frost and strong winds and also for irrigation scheduling.Prediction and early warning with good lead times are vital not only for enhancing food and agricultural production, butalso in the utilization and management of fresh water, energy and other natural resources that are sensitive to extremeweather and climate events.

Over the past years, WMO has been assisting Member countries through its Agricultural Meteorology Programmeto enhance the application of science and technology for improved agricultural production. The provision of scientificallybased forecasts and warnings, as well as improved agrometeorological information and services, has enabled every nationin the world to forewarn and protect their communities from the threat of tropical cyclones, floods, droughts anddesertification, locust invasion, forest fires, severe storms and other weather-induced natural disasters.

Given the significance of the impact of extreme meteorological and hydrological events on agriculture, theCommission for Agricultural Meteorology (CAgM) at its eleventh session held in Havana in February 1995 invitedDr H.P. Das (India/Chairman), Dr K. Anaman (Australia), Ms T.I. Adamenko (Ukraine), Ms B. Cusursuz (Romania),Dr G. Johnson (United States) and Mr R.G. Gommes (FAO) to serve as members of a working group to survey andsummarize existing knowledge on the application of agrometeorological information needed to better cope with extremeevents. The group was also asked to provide examples of the operational use of such information for Member countries andto prepare guidance material for training purposes; to coordinate the design and establishment of a database of extremeevents which have significant social and economic impacts on agriculture, foresty and fisheries; and to examine themethods used for assessing the economic impacts of extreme meteorological events on agricultural production.

The report of the group addresses these issues which are of great interest to all nations and in particular to thedeveloping countries. I believe that the publication of this report as a WMO Technical Note will contribute to thedevelopment of sustainable agricultural strategies by WMO Members. It is therefore with much pleasure that I take thisopportunity to express the gratitude of the World Meteorological Organization to all members of the working group whocontributed to the report.

(G.O.P. Obasi)Secretary-General

FOREWORD

This technical note demonstrates the effects of extreme meteorological events on agricultural production and summarizesexisting knowledge on the application of agrometeorological information needed to better cope with extrememeteorological events.

Chapter 1 (Introduction) deals with the definition and type of extreme climatic events with emphasis on therequirements of an adequate database to estimate the risk of extreme events in quantitative terms. The impact of climaticextremes on agricultural production and the socio-economic consequences are discussed in general with a focus on disasterprevention, preparedness and rehabilitation.

Chapter 2 (Agrometeorological aspects of drought and desertification) deals with the concept, definitions and causes ofdrought and desertification. Spatial and temporal aspects of drought and their impact on agriculture are explained withparticular reference to socio-economic implications. Forecasting, monitoring, early warning of drought and its adaptationand management are well documented in the context of agricultural production. Distribution and trends of desertification,with particular reference to agrometeorological aspects, are also included, along with the methods of its control, monitoringand prevention.

Chapter 3 (Incidence, prediction, monitoring and mitigation measures of tropical cyclones and storms surges) describes thegeographical distribution and related characteristics of tropical cyclones with emphasis on prediction, monitoring andmitigation aspects. Storm surges, often associated with tropical cyclones, are also discussed. Agrometeorological lossesassociated with some of the worst cyclones are illustrated with examples. The chapter ends with a description of cyclonewarning systems and disaster management.

Chapter 4 (Assessing the economic and social impacts of extreme events on agriculture and use of meteorologicalinformation to reduce adverse impacts) looks at the economic and social costs extreme events can have in relation toagriculture and agricultural communities. Data, information and services are considered as economic resources for analysisand discussed at length.

Chapter 5 (Assessing the impact of extreme weather and climate events on agriculture, with particular reference to floodingand heavy rainfall) presents a synopsis of responses to a survey on extreme weather and climate events. The impacts offlooding and heavy rainfall on agriculture are brought out in this chapter.

Chapter 6 (Hail, high winds and cold injury) highlights the effects of some local extreme weather events onagriculture. It shows how hail, high winds and extreme cold weather, including cold injury, affect agricultural productionin Ukraine.

Chapter 7 (Locusts) describes some principal pests including locusts, grasshoppers and army worms, the devastationthey cause to crops and how their migration is largely controlled by agrometeorological, climatological and synopticconditions.

Chapter 8 (Specification for a database of extreme agrometeorological events) is an attempt to design a specification fora database of extreme agrometeorological events. It provides the component, methodology and structure of a database ofagricultural disasters resulting from extreme geophysical and man-made factors with an atmospheric component.

Chapter 9 (Conclusions and recommendations) highlights the impact extreme agrometeorological events have aroundthe world. Recommendations regarding information systems; monitoring, early warning and remedial measures; trainingand awareness raising; and global cooperation are provided.

Noting that extreme agrometeorological events continue to occur in many parts of the world with negative impactson agricultural production, it is recommended that the Commission for Agricultural Meteorology continues to study thissubject with renewed terms of reference.

SUMMARY

RÉSUMÉ

La présente note technique confirme les effets des phénomènes météorologiques extrêmes sur la production agricole et faitle point sur les connaissances actuelles concernant l’application de l’information agrométéorologique nécessaire pour mieuxfaire face à ces phénomènes.

Le chapitre 1 (Introduction) traite de la définition et du type des phénomènes climatiques extrêmes, et en particulierde la nécessité de disposer d’une base de données convenable pour évaluer les risques de tels phénomènes sur le planquantitatif. L’incidence des extrêmes climatiques sur la production agricole et les conséquences socio-économiques sontexaminées d’un point de vue général, l’accent étant mis sur la prévention des catastrophes, la préparation à ces catastropheset la réhabilitation.

Le chapitre 2 (Aspects agrométéorologiques de la sécheresse et de la désertification) porte sur le concept et la définitionde la sécheresse et de la désertification ainsi que sur leurs causes. Les éléments spatiaux et temporels de la sécheresse et leurincidence sur l’agriculture sont examinés, notamment du point de vue des conséquences socio-économiques. Les aspectsrelatifs à la prévision, à la surveillance, à l’annonce précoce et à la gestion des phénomènes de sécheresse ainsi qu’àl’adaptation à leurs effets sont largement abordés, toujours dans le contexte de la production agricole. Les questions de larépartition et de l’évolution de la désertification, eu égard en particulier aux aspects agrométéorologiques, sont égalementabordées, de même que celles qui concernent les méthodes employées pour lutter contre ce phénomène et en assurer lesuivi et la prévention.

Le chapitre 3 (Incidence, prévision et surveillance des cyclones tropicaux et des marées de tempête et mesures d’atténuationde leurs effets) décrit la distribution géographique et les caractéristiques connexes des cyclones tropicaux, l’accent étant missur la prévention et la surveillance de ces phénomènes et sur l’atténuation de leurs effets. Les marées de tempête, donts’accompagnent souvent les cyclones tropicaux, sont également mentionnées. Les pertes agrométéorologiques dues àquelques-uns des cyclones les plus dévastateurs sont illustrées par des exemples. Le chapitre se termine par une descriptiondes systèmes d’avis de cyclone et par des considérations sur la gestion des catastrophes.

Le chapitre 4 (Evaluation des répercussions économiques et sociales des phénomènes extrêmes sur l’agriculture etutilisation de l’information météorologique pour en atténuer les effets néfastes) porte sur les coûts économiques et sociauxque peuvent avoir les phénomènes extrêmes en ce qui concerne l’agriculture et les communautés agricoles. Les données, lesinformations et les services sont considérés comme des ressources économiques aux fins d’analyse et font l’objet d’unexamen approfondi.

Le chapitre 5 (Evaluation de l’incidence des phénomènes météorologiques et climatiques extrêmes sur l’agriculture,notamment pour ce qui concerne les inondations et les fortes pluies) présente un résumé des réponses obtenues dans le cadred’une enquête sur les phénomènes météorologiques et climatiques extrêmes. On attire en particulier l’attention sur lesconséquences des inondations et des fortes pluies pour l’agriculture.

Le chapitre 6 (Grêle, vents forts et dégâts dus au froid) insiste sur les effets de certains phénomènes météorologiquesextrêmes sur l’agriculture. On y décrit la manière dont la grêle, les vents forts et les périodes de froid extrême (y compris lesdégâts dus au froid) ont des effets négatifs sur la production agricole en Ukraine.

Le chapitre 7 (Criquets) donne la description de certains des principaux ravageurs (criquets, sauterelles, chenilleslégionnaires, etc.) et des dégâts qu’ils causent aux cultures et précise le rôle déterminant que jouent les conditionsagrométéorologiques, climatiques et synoptiques dans leur migration.

Le chapitre 8 (Spécifications d’une base de données sur les phénomènes agrométéorologiques extrêmes) est une tentatived’élaboration de spécifications pour une base de données sur les phénomènes agrométéorologiques extrêmes. On y préciseles composantes, la méthodologie et la structure d’une base de données sur les catastrophes agricoles qui résultent defacteurs géophysiques et anthropiques extrêmes et qui présentent un élément atmosphérique.

Le chapitre 9 (Conclusions et recommandations) insiste sur les répercussions qu’ont les phénomènesagrométéorologiques extrêmes dans le monde entier. On y formule en outre des recommandations concernant les systèmesd’information; les systèmes de surveillance et d’alerte précoce et les mesures correctives ; la formation et la sensibilisation;et la coopération à l’échelle du globe.

Etant donné que les phénomènes agrométéorologiques extrêmes continuent d’avoir des effets néfastes sur laproduction agricole dans de nombreuses régions du globe, il est recommandé que la Commission de météorologie agricolecontinue d’étudier cette question dans le cadre d’un mandat renforcé.

РЕЗЮМЕ

В настоящей технической записке демонстрируется воздействие экстремальных метеорологических явлений на

сельскохозяйственное производство и содержится резюме существующих знаний по применению

агрометеорологической информации, необходимой для того, чтобы лучше противостоять экстремальным

метеорологическим явлениям.

Глава 1 (Введение) содержит определения и описание видов экстремальных климатических явлений с

упором на потребности в адекватной базе данных для оценки риска экстремальных явлений в количественном

выражении. Излагаются также вопросы воздействия экстремальных климатических явлений на

сельскохозяйственное производство и их социально-экономические последствия в целом с упором на

предотвращение последствий стихийных бедствий, подготовку к ним и устранение ущерба.

Глава 2 (Агрометеорологические аспекты засухи и опустынивания) рассматривает концепцию определения

и причин засухи и опустынивания. Пространственные и временные аспекты засухи и воздействие засух на

сельское хозяйство освещается с особым упором на социально-экономические последствия. Прогнозирование,

мониторинг, заблаговременное предупреждение засухи и адаптация к условиям засухи и борьба с ней хорошо

излагаются в контексте сельскохозяйственного производства. Вопросы распространения и тенденции

опустынивания с особым упором на агрометеорологические аспекты также раскрываются в этой главе наряду с

методами борьбы с засухой, мониторинга и предотвращения засухи.

Глава 3 (Появление, предсказание, мониторинг и меры по смягчению последствий тропических циклонов и

штормовых нагонов) описывает географическое распространение и соответствующие характеристики

тропических циклонов с упором на прогноз, мониторинг и аспекты смягчения последствий. Излагаются также

вопросы, касающиеся штормовых нагонов, которые часто связаны с тропическими циклонами. Например,

иллюстрируются потери в плане ухудшения агрометеорологических условий в связи с некоторыми самыми

сильными циклонами. Глава заканчивается описанием систем предупреждений о циклонах и изложением

вопросов обеспечения готовности к стихийным бедствиям.

Глава 4 (Оценка экономических и социальных последствий экстремальных явлений для сельского хозяйства и

использование метеорологической информации для уменьшения негативных воздействий), в которой

рассматриваются социально-экономические потери в финансовом плане от экстремальных явлений, которые

могут иметь отношение к сельскому хозяйству и сельскохозяйственным общинам. Данные, информация и

обслуживание рассматриваются как экономические ресурсы для анализа, и эти вопросы излагаются достаточно

подробно.

Глава 5 (Оценка воздействий экстремальных погодных и климатических явлений на сельское хозяйство с

особым упором на затопления и ливневые осадки) представляет собой совокупность ответов на обзор по

экстремальным погодным и климатическим явлениям. В этой главе рассматривается воздействие затоплений и

ливневых осадков на сельское хозяйство.

Глава 6 (Град, сильные ветры и повреждения низкими температурами) освещает воздействие некоторых

локальных экстремальных погодных явлений на сельское хозяйство. В ней описывается то, каким образом град,

сильные ветры и экстремально холодная погода, включая повреждения низкими температурами, влияют на

сельскохозяйственное производство на Украине.

Глава 7 (Саранча) описывает некоторых основных вредителей, включая саранчу, кузнечиков и гусениц,

опустошительный ущерб, который они могут наносить сельскохозяйственным культурам, и то, каким образом их

миграция в основном обусловлена агрометеорологическими, климатологическими и синоптическими условиями.

Глава 8 (Спецификация базы данных экстремальных агрометеорологических явлений) представляет собой

попытку составить спецификацию для базы данных экстремальных агрометеорологических явлений. В ней

содержится описание методологии и структуры базы данных сельскохозяйственных бедствий, являющихся

результатом экстремальных геофизических и антропогенных факторов, включая атмосферный компонент.

Глава 9 (Выводы и рекомендации) освещает воздействие экстремальных агрометеорологических явлений по

всему земному шару. Содержит рекомендации, касающиеся информационных систем, мониторинга,

заблаговременного предупреждения и мер по устранению последствий; обучения и повышение уровня

осведомленности; и касается также вопросов глобального сотрудничества.

Принимая во внимание, что экстремальные агрометеорологические явления будут продолжать иметь

место во многих частях земного шара с негативными последствиями для сельскохозяйственного производства,

рекомендуется, чтобы Комиссия по сельскохозяйственной метеорологии продолжала изучать эти вопросы при

обновленном круге своих обязанностей.

RESUMEN

En esta nota técnica se demuestran los efectos de los fenómenos meteorológicos extremos sobre la producción agrícola y seresumen los conocimientos existentes acerca de la aplicación de la información agrometeorológica necesaria para poderhacer frente más eficazmente a los fenómenos meteorológicos extremos.

En el Capítulo 1 (Introducción) se abordan la definición y el tipo de los fenómenos climáticos extremos, prestandoespecial atención a los requisitos de una base de datos adecuada para la estimación de los riesgos de los fenómenos extremosen términos cuantitativos. Se examinan en general los efectos de los extremos climáticos sobre la producción agrícola y susconsecuencias socioeconómicas, prestando especial atención a la prevención de desastres, las medidas de preparación y larehabilitación.

El Capítulo 2 (Aspectos agrometeorológicos de la sequía y la desertificación) trata del concepto, las definiciones y lascausas de la sequía y la desertificación. Se explican los aspectos espaciales y temporales de la sequía y sus efectos sobre laagricultura, con especial referencia a las implicaciones socioeconómicas. La predicción, la vigilancia, la alerta temprana dela sequía y las medidas de adaptación y gestión están bien documentadas en el contexto de la producción agrícola. Seincluyen también la distribución y las tendencias de la desertificación, con particular referencia a los aspectosagrometeorológicos, así como los métodos de control, vigilancia y prevención.

En el Capítulo 3 (Incidencia, predicción, vigilancia y medidas de mitigación de los ciclones tropicales y de las mareas detempestad) se describen la distribución geográfica y las características conexas de los ciclones tropicales, poniéndose énfasisen los aspectos de predicción, vigilancia y mitigación. Se estudian también las mareas de tempestad que a menudoacompañan los ciclones tropicales. Se presentan casos ilustrativos de las pérdidas agrometeorológicas asociadas con algunosde los ciclones que han causado mayores destrozos. El capítulo concluye con una descripción de los sistemas de alertaciclónica y gestión de desastres.

En el Capítulo 4 (Evaluación de los efectos económicos y sociales de los fenómenos extremos en la agricultura y lautilización de la información meteorológica para reducir los impactos adversos) se examinan los costos económicos y sociales delos fenómenos extremos en relación con la agricultura y las comunidades agrícolas. Los datos, la información y los serviciosse consideran recursos económicos para el análisis y son objeto de un estudio pormenorizado.

El Capítulo 5 (Evaluación de los efectos de los fenómenos meteorológicos extremos y de los fenómenos climáticos en laagricultura, con particular referencia a las crecidas y a las precipitaciones intensas) presenta una sinopsis de las respuestas a unaencuesta sobre fenómenos meteorológicos y climáticos extremos. En ese capítulo se abordan los efectos de las inundacionesy de las precipitaciones intensas sobre la agricultura.

En el Capítulo 6 (Granizo, vientos fuertes y daños causados por el frío) se subrayan los efectos de algunos fenómenosmeteorológicos extremos para la agricultura. Se explica cómo el granizo, los vientos fuertes y el frío extremo, incluidos losdaños causados por el frío, afectan la producción agrícola en Ucrania.

El Capítulo 7 (Plagas de langostas) describe algunas de las principales plagas, incluidas las langostas, los saltamontesy las orugas negras, sus devastadores efectos sobre los cultivos y la manera en que su migración está controlada mayormentepor las condiciones agrometeorológicas, climatológicas y sinópticas.

En el Capítulo 8 (Especificaciones para una base de datos sobre fenómenos agrometeorológicos extremos) es un esbozo dediseño de una especificación para una base de datos sobre fenómenos agrometeorológicos extremos. Se presentan loscomponentes, la metodología y la estructura de una base de datos sobre desastres agrícolas que obedecen a factoresgeofísicos extremos y antropogénicos con componente atmosférico.

El Capítulo 9 (Conclusiones y recomendaciones) subraya las repercusiones de los fenómenos agrometeorológicosextremos en todo el mundo. Se presentan recomendaciones relativas a los sistemas de información, vigilancia, alertatemprana y medidas correctivas; capacitación y elevación de la concienciación, y cooperación a escala mundial.

Teniendo presente que en muchas partes del mundo siguen ocurriendo fenómenos agrometeorológicos extremos quetienen efectos negativos en la producción agrícola, se recomienda que la Comisión de Meteorología Agrícola continúeestudiando estos temas con un mandato renovado.

Dr H.P. DasAgricultural Meteorology DivisionIndia Meteorological DepartmentPune 411 005INDIA

Dr K.A. AnamanDepartment of EconomicsUniversity of Brunei DarussalamBandar Seri Begawan 2028BRUNEI DARUSSALAM

Mr R.G. GommesCoordinatorAgrometeorology Group/SDRNFAOVia delle Terme di Caracalla00100 ROMEITALY

Dr G. JohnsonUS Department of AgricultureNatural Resources Conservation ServiceNational Water and Climate CenterPortland, Oregon 97204-3224UNITED STATES

Ms T.I. AdamenkoAgrometeorological DivisionState Committee for Hydrometeorology6 Zolotovorotskaya Street252601 MSP34 KievUKRAINE

LIST OF CONTRIBUTORS

CHAPTER 1

INTRODUCTION(by H.P. Das)

1.1 DEFINITION OF EXTREME EVENTS

Extreme events are infrequent meteorological phenomena that surpass a definedthreshold. The perceived severity depends on the vulnerability of the natural environment and human society to that event. This implies that the definition of an extreme event can depend strongly on location. Susman, et al.(1983) described a disaster as “the interface between an extreme physical event anda vulnerable human population”. In the same way, an extreme agrometeorologicalevent is the interaction between a vulnerable agricultural system and extremeweather conditions. However, the definition of extreme agrometeorological eventsis broader, as it also includes weather conditions conducive to the development ofagents such as pests and diseases that adversely affect all aspects of agricultureincluding livestock and pasture, forests and fisheries (Gommes, 1997).

One important aspect of extreme events is the apparent randomness andabruptness with which they arrive. Global changes in air pollution, acid deposition,desertification, water shortages, salt water intrusion and soil degradation are alsoserious but tend to arrive slowly enough that regional, national and local authoritiescan adopt successful long-term counter-measures.

1.2 TYPES OF EXTREME CLIMATIC EVENTS

Generally, plants exhibit particular thresholds for the various climatic variablesdetermining plant growth and development. The optimum values for plant growthmay not necessarily be those for its development. The climatic events whichadversely affect agricultural production may be linked to an extreme value of oneparameter or another. Some of the important extreme climatic events from an agriculture and livestock point of view are:

(a) Tropical storms (cyclones, hurricanes, typhoons, etc.) associated with high winds,flooding and storm surges;

(b) Floods (other than those related to tropical storms), heavy rains during monsoonand waterlogging;

(c) Severe thunderstorms, hailstorms, tornadoes and squalls;(d) Drought and heatwaves;(e) Cold spell, low temperature, frost, snow and ice storms;(f) Dust storms and sand storms;(g) Weather conducive to fires (lightning); and(h) Weather encouraging pests and diseases of crops and livestock.

Besides the above climatic events, the following geological/geophysical extremeevents are also hazardous to society at large:

(a) Volcanic eruptions;(b) Earthquakes and tsunamis;(c) Avalanches; and(d) Landslides and mudslides.

1.3 AGROMETEOROLOGICAL DATA RELATED TO EXTREME EVENTS

The first and the most basic requirement in agrometeorological hazard assessmentfor extreme events is an adequate database. If sufficient quality data are available,

it may be feasible to estimate the risk of the extreme events and their damage inquantitative terms. An overall assessment will include information not only onmeteorological and hydrological aspects but also on social, economic, geographicaland other factors. The evaluation of risk or the investigation of disaster potentialcan generally be done in a fairly straightforward way, using the long-term meteorological and hydrological records of the country, supplemented, if necessary,by data available from neighbouring countries. Analysis of these data may be helpful in making decisions on social, economic, regional and other considerationsand to plan and organize protective measures.

As a first step, the climatological records should be analyzed in order to discoverhow often an extreme event, say tropical cyclones of various intensities, strikedifferent areas or regions of the country. All possible sources should be considered inorder to build up the most complete data possible. Old records in libraries andnewspaper offices often furnish valuable information.

While measures to optimize an extreme events database should be vigorouslypursued, it must be recognized that basic data collection, processing and storage,remain the cornerstone of any research and of operational aspects of extreme events.For each extreme event, the database should include location, time and detailsabout the severity of the phenomenon and the extent of damage or injury, preferablyquantitatively.

Accurate information on extreme meteorological events is extremely importantto farmers in maximizing their production. The information is also useful formodifying the crop environment, protection from frost and strong winds and alsofor irrigation scheduling. The extension of cultivation into less suitable climatesincreases the risk of damage by the climate, particularly meteorological extremes.The successful development of a country’s agricultural economy is, therefore, to alarge extent, dependent on the use of climatic information, especially on adverseagrometeorological factors. To examine the effect of agroclimatic extreme events,all aspects of the climatology of the locality must be considered.

Observation systems must be devised in anticipation of damage from theextreme weather. The nature of the observation for each extremeagrometeorological event will vary with the type of hazard. A country prone totropical cyclones should install additional observation facilities to supplement thebasic meteorological network used for its normal forecasting and climatologicalpurposes as has been done in India. The data collected should include the loss ofhuman and cattle lives, the number of people injured, areas inundated and/ordamaged, crop losses, the number of dwellings destroyed and damaged etc.

The real-time monitoring and assessment of drought requires collection ofrainfall and other related drought data. It is necessary to supplement the synopticdata with information on evaporation, radiation, soil moisture, the undergroundwater table etc. The accumulated precipitation amount is one of the most essentialelements in a real-time drought surveillance service and must supplement data ontemperature, humidity, cloudiness and wind. In relation to weather hazards, data arealso required on the state and stage of crops.

Data including frequencies and duration of water levels and dischargesexceeding certain thresholds are very important for design and planning of theobservation system. Usually the regular observation network does provideinformation on storm rainfall distribution and on flood peak discharges of tributarystreams. During severe floods, permanent stream gauge installations can be washedaway and records lost. For such reasons valuable information can be obtained by aflood survey conducted in the storm/flood area following a severe occurrence. Soilmoisture data at weekly/monthly intervals, presented as graphs, are useful inscheduling the application of irrigation and also in river forecasting. Data on extent,depth and water equivalent of snow cover together with the daily and accumulatednumber of degree days above or below a certain base and melting degree days areuseful for forecasting snow-melt run-off. A particularly important factor on which tohave information is frost which can be very damaging to plants.

For crops affected by pests and diseases, information is needed of the state andstage of the crop, the availability and release of spores, incidence and spread of

2 AGROMETEOROLOGY RELATED TO EXTREME EVENTS

infection, etc. Information is also required on the hatching of various insects, thebuild up of insect populations or their invasion from other territories.

Since there is a direct relationship between weather and fire danger, andbetween weather and fire behaviour, a knowledge of past, present and future weatheris desirable. This should include temperature, relative humidity, wind, precipitationand thunderstorm data. Information is also required on the state of forest litter andits liability to burn.

The ability of cattle in the open air to withstand low temperatures is fairlystrong. However, it is the secondary effect of weather often accompanying a coldwave which causes widespread livestock losses. Snow covers forage and drinkingwater supplies freeze. As a result cattle caught in a winter storm can starve ratherthan die directly from the cold temperatures. Cattle, pigs, poultry and otherlivestock are adversely affected by high temperature together with high relativehumidity. Meteorological data on these aspects are very useful in forecasting extremeepisodes and minimizing losses.

Without doubt, the most spectacular observational tool of the last few decadeshas been the meteorological satellite. Some satellites provide data on “wetness” ofthe vegetated surface and “surface wetness”. Such data, though not beingagrometeorological extreme events can provide useful information as to when ameteorological extreme event such as a tropical storm, excessive rains, drought or anattack by pests may occur and are thus highly useful.

1.4 EXTREME EVENTS AND AGRICULTURAL PRODUCTION

Agriculture depends on the mean climate of a particular region. Each plant has itsown climatic requirements for growth and development and any large-scale deviation from them exerts a negative influence.

As the temperature of the atmosphere and of the soil varies, so the developmentrate of plants varies up to an optimum value beyond which it tends to decrease.Plant growth is most sensitive to temperatures just above a threshold value and nearthe maximum value where growth normally stops. Therefore, periods of extremetemperature values, which are well below the threshold value or very high abovethe maximum value are hazardous to plant development and growth. Periods ofextreme temperature conditions such as those experienced during extreme coldspells causing cold stress and frost, or high temperatures and heat waves leading toheat stress can affect agricultural production. Snow and ice storms in late spring orearly autumn are very hazardous to many temperate crops, exposing them to layers ofsnow and ice and causing freezing of the crop.

Similarly, extremes in moisture conditions, namely dry desiccating winds,drought episodes and low moisture availability as well as very humid atmosphericconditions including wet spells affect agriculture. High soil moisture in situations ofwaterlogging and flooding associated with heavy rainfall and tropical stormsadversely effects plant growth and development since it influences the rate oftranspiration, leaf-area expansion and, ultimately, plant productivity. Drasticchanges in rainfall distribution can have a very significant impact, particularly inclimatically marginal zones such as arid, semi-arid and sub-humid areas where theincidence of widespread drought is frequent.

There are, however, some advantages to dry spells or drought at certain times inthe development of some crops such as sugar cane where a brief dry spell is essentialduring the pre-harvest stage. This helps to concentrate or increase the sucrosecontent of the cane. Additionally, there is often a lower incidence of pests anddiseases in periods of drought.

In regimes prone to strong winds, damage to plants and reduced agriculturalproduction occur as a result of very high evapotranspiration (ET) rates. Strong windscause mechanical damage or breakage to herbaceous plants with weak stems such assugar cane and banana. Windstorms and tropical storms (hurricanes and typhoons)with very high winds can destroy fields of cereals within minutes reducing the yieldsignificantly.

CHAPTER 1 — INTRODUCTION 3

Extreme climates also impact on plants through the development of pests anddiseases. For example, the growth and development of locusts and grasshoppersdepend heavily on the climatic conditions in the infected areas. Locusts cannotthrive under severe cold conditions or extremely wet situations.

Other hazardous climatic events include thunderstorms, tornadoes, squall lines,hailstorms and weather-related wildland fires.

1.5 SOCIO-ECONOMIC IMPACT OF EXTREME EVENTS

As populations grow, more people become vulnerable to damage from the occurrence of extreme events in nature. Social losses from avalanches, earthquakes,tropical cyclones and many other natural hazards are increasing. This is the caseeven though new measures for dealing with hazards proliferate. In some areasmeasures are curbing losses to a significant extent. Generally, sophisticated meansof providing relief in times of disaster are better developed than the means of preventing disasters.

People who have experienced tropical cyclones are generally very receptive toany warnings that are issued and to follow advice given, including instructions forevacuation to safer areas. Everyone should be made aware of the dangers posed bytropical cyclones. Furthermore, since memories are apt to fade, this awareness mustbe kept alive and up-to-date even for those who have experienced a tropical cyclonein the past. Loss of life as a result of storm warnings being disregarded is aconsequence of people’s attitudes and emotions. Undoubtedly, this is a problemdeserving close attention.

In many respects, the human response to the threat of danger from extremeevents is the very core of disaster prevention and preparedness. Ultimately, thesuccess or failure of the warning systems depends upon people. An accurate forecast,a well-designed disaster preparedness system and all the aids that technology canprovide count for little if societal response is not in tune with the realities of the event.

The impact of extreme events on society can be positive or negative. Negativeor adverse impacts include damage or loss due to droughts, tropical cyclones andfloods. However, sometimes extreme events have positive effects such as increasedrainfall in coastal areas from tropical cyclones, fixing of atmospheric nitrogen bythunderstorms, germination of native plant species resulting from bush fires, siltdeposition, water reserves repletion and soil desalinization due to floods. Ofparticular note in this context are river bed changes and major landslides which maycompletely modify the agricultural landscape.

Extreme events can be direct or indirect in their effect. Direct impacts arisefrom the direct physical contact of the event with people, their animals and theirproperty. For example, tropical cyclones directly cause the loss of farmers’ standingcrops and damage irrigation facilities; drought directly reduces crop yields and leadsto the death of livestock and people. Indirect effects tend to appear progressively asa result of low incomes, decreases in production, environmental degradation andother factors related to the disaster. Indirect impacts include the evacuation ofpeople in the event of cyclone landfall, disruption to households, stress inducedsickness and apprehension, (Handmer and Smith, 1992; Anaman, 1996) and tidalwave-related salinization of soils. Factories and warehouses may be out ofcommission for a time. In agriculture there can be large losses in primary productionon account of delays in the recovery of arable land that has been inundated.

Extreme events cause many losses of a personal and domestic nature. The loss ofpersonal belongings, such as clothing, furniture and household items, can be a severeblow to families whose financial reserves are small. The breakdown in public utilitiescan lead to considerable losses in the domestic context. For example, an electricityfailure which puts refrigerators out of action causes perishable foodstuffs to bewasted. All these losses, when aggregated, can amount to a substantial financial lossfor a whole community.

The emotional shock of disaster, the death or injury of family members, theseparation of families, changes in living accommodation, the burden of hardship


from material losses, physical handicap resulting from injury and the loss of incomeor employment can all create socio-economic problems and affect the ability of anindividual or family to recover. The agencies concerned must be conscious of theneed to deal with such problems. Some situations require only counselling andadvice but there can be many cases in which material help and constant support areneeded.

1.6 PREVENTION AND PREPAREDNESS

In recent decades people throughout the world have become increasingly alarmedby natural disasters which are becoming more frequent and more destructive. Theforces of nature cannot, yet, be controlled. Humans cannot prevent the formationof a tropical cyclone, an earthquake or the eruption of a volcano. However, we areable to contain rivers, stem tides and build structures that give considerable, if nottotal, resistance to the forces of nature. Since natural phenomena will continue tooccur, the problems they present must be faced, and due priority to policies for disaster planning, preparedness and prevention must be given(ESCAP/WMO/LRCS, 1977).

Disaster prevention measures are complex because of their wide scope and theirtechnical nature. They relate not merely to the disasters themselves but also reflectthe interaction between development and the environment on the one hand andbetween social and economic interests on the other. Except where socialconsiderations are an overriding priority, decisions on disaster prevention should bebased on cost-benefit and associated criteria. For example, a proposal to locate anindustry in a disaster-prone area should be examined in relation to the probability ofdamage (vulnerability) and economic factors such as access to water, energy,transport, labour, raw material, etc. The environmental impact of disasterprevention measures should also be considered. Flood control measures and floodmanagement may yield valuable benefits by reducing risks of silting, soil erosion andlandslide. When considering the social, economic and even psychological factors atnational, regional and local levels, the complexities of disaster prevention and rangeof technical options to employ are great.

Disaster preparedness is the plan of action or emergency measures which comeinto force when an extreme agrometeorological event is about to occur. Thesemeasures remain in force until some time after the adverse conditions have abated,because action is required not only when an event is approaching but also when it isactually present and in its aftermath.

In an integrated disaster plan, there are two categories of measures. The firstconcerns those of a permanent nature, referred to as prevention measures. Theseinclude structural components – levees, dams, reservoirs, etc. – and non-structuralcomponents – land use and zoning, building codes, etc. The second category –preparedness measures – consists of emergency measures, though these must also beplanned well in advance. Both categories are essential and should not be viewed asseparate undertakings but as essential parts of the overall system for protecting lifeand property.

1.7 REHABILITATION

If, as a result of the material damage suffered in a locality a large-scale programmeof rehabilitation is required, the aim might be to improve rather than merely restoreexisting living standards and social conditions. Morale is a key factor in rehabilitation.It is possible for people to emerge from a disaster feeling hopeless and apathetic. Ifthis attitude is allowed to persist, people will become over-dependent on welfareservices and be a permanent burden to the nation. High morale can be fostered byhelping people but at the same time promoting self-reliance.

Rehabilitation should be carried out via a two-pronged programme coveringboth the victims of the disaster and the public services and amenities

CHAPTER 1 — INTRODUCTION 5

(ESCAP/WMO/LRCS, 1977). For the victims, assistance may include the repair ofhomes, the provision of basic home needs such as furniture and kitchen utensils, theprovision of food and clothing and resettlement. In the agricultural sector, allpossible help should be directed at the recovery of land, resowing, desalination,replacement of crops and livestock, repair of irrigation facilities, etc. The costs ofrestoring these facilities can be very heavy and this consideration should becompared with the area’s vulnerability and other factors.

• Anaman, K.A., 2003: Assessing the economic and social impacts of extreme eventson agriculture and use of meteorological information to reduce adverse impacts. Draftcontribution to working group report on agrometeorology related to extremeevents, WMO, Geneva.

• ESCAP/WMO/LRCS, 1977: Guidelines for disaster prevention and preparedness intropical cyclone areas. Geneva/Bangkok.

• Gommes, R.G., 1997: Extreme agrometeorological events. CAgM Report No. 73,WMO, Geneva.

• Handmer, J. and Smith, D.I., 1992: Cost-effectiveness of flood warnings. Reportprepared for the Australian Bureau of Meteorology by the Centre for Resource andEnvironment Studies. Australian National University, Canberra, Australia,pp. 50.

• Susman, P., O’Keefe, P., Wisher, B., 1983: Global disasters, a radical interpretation. In: Hewitt, K. (ed.), Interpretations of calamity, Risks and HazardsSeries: 1, Allen and Unwin Inc., Boston, pp. 263–280.


REFERENCES

CHAPTER 2

AGROMETEOROLOGICAL ASPECTS OF DROUGHT ANDDESERTIFICATION(by H.P. Das)

2.1 DROUGHT

Since the beginning of human civilization drought has had severe and sometimescatastrophic effects on vital human activities around the world. Among the varioushazards of nature, drought is one of the most disastrous. Few parts of the world canboast of not having faced this calamity at some stage in history. Drought is a progressive (“creeping”) phenomena. Both the onset and the end of a drought canoften be difficult to identify because they lack a sharp distinction from non-droughtdry spells. Drought creates innumerable problems immediately or with a time lag asthe economy gradually experiences the adverse effects of this phenomenon. If thedrought is widespread and prolonged, the cumulative effect is usually disastrous. Amajor drought not only causes serious dents in the economy but also dampens peoples’ resolve. Drought impacts are felt not only on agriculture but also on urbanwater supplies, industrial production, pollution control, navigation and energyrecreation.

The occurrence of severe droughts throughout Africa and in India, NorthAmerica, China, the Soviet Union, Australia, and Western Europe in the 1980sonce again underscored the vulnerability of both developed and developing societiesto drought. Even in the present age of high technology and instant communication,agricultural and livestock production can be sharply reduced by drought-relatedstresses. No country can claim to be immune from the uncertainties of seasonal orannual rainfall. Although we have little capability (if any) to avert meteorologicaldrought, reliable information about drought and its impacts could be used byindividual farmers as well as planners and political leaders at national level toimprove society’s ability to minimize the scope and severity of its consequences.Societal vulnerability to drought is increasing, largely because of population growthand society’s increasing demand and competition for limited water resources.

Drought has also been blamed for prompting mass migrations, environmentaldegradation (often referred to as desertification) and internal unrest. While droughtby itself may not appear to be a major cause of societal dislocation, it can combinewith underlying societal problems to initiate new changes or to accelerate theotherwise slower changes that are already underway. Often the impacts of droughtlingers long after a drought has ended, thereby dissociating the drought itself frommany of its impacts.

Droughts often become highly visible when they are associated with famine. Thetruth is that governments prefer to blame natural factors such as droughts forextreme food shortages. For the most part, droughts can occur without precipitatinga famine situation; and historical records have shown that famines have frequentlytaken place in the absence of drought conditions. Many authors have explainedwhy droughts need not result in famine and famines do not necessarily have theirorigin in drought (Sen, 1981; Watts, 1983; Torry, 1984). Often, drought, a “creeping” phenomenon, combines with other underlying societal and environmental conditions to produce famine-like conditions. The fact that droughtneed not be a causative agent in producing famine became clear in 1992. Thereason 1992 is cited as the critical year for understanding famine comes from thesituations of three distinct groups of people (Bosnians, Somalis, and Kurds) in threeseparate parts of the world (Europe, Africa and the Middle East, respectively). Ineach case one found starving people, yet in each case, weather played little or no

2.1.2DROUGHT AND FAMINE

2.1.1INTRODUCTION

role, while military conflict played a major one. Though governments would preferto have blamed their inability to feed their people on weather-related problems,these three 1992 famine situations do not allow them to get away with it (Glantz,et al., 1985).

In order to discuss drought phenomena and their impacts, the concept “drought”needs to be defined. The definition and particularly the quantification of the term“drought” (and even the concept of drought) is controversial. In the past, there hasbeen no universally accepted definition of drought, partly perhaps because the concept is not absolute but relative to users and expectations. In very general terms,drought is a condition of moisture deficit of sufficient magnitude to have an adverseeffect on vegetation, animals and people over a sizeable area (Warrick, 1975).Drought has been grouped by type as follows: meteorological, hydrological, agricultural and socio-economic (Wilhite and Glantz, 1985).

Meteorological drought can be defined as a percentage departure from the longterm average rainfall in a given region. A meteorological drought is sometimesdifficult to identify with any degree of reliability, because meteorological andclimatological information in many countries is either not available at all oravailable for only short time periods or is of relatively poor quality. Definitions ofmeteorological drought are considered as region specific since the atmosphericconditions that result in deficiencies of precipitation are highly variable from regionto region. Human perceptions of these conditions are equally variable. Hydrologicaldrought is represented by the water shortage formed by an imbalance betweensurface water and underground water. It is mainly affected by hydrological factorssuch as surface run-off and shallow or deep drainage. Meteorological drought, ifprolonged, could result in hydrological drought with a marked depletion of surfacewater and consequent drying up of reservoirs, lakes, streams and rivers and a fall inthe water table. Hydrological drought is often out of phase with meteorologicaldrought.

In defining agricultural drought, rainfall deficiency has to be taken into account,along with the physical and biological aspects of plants, interactions within the soil-plant-atmosphere continuum and the balance between the water demand of plantsand its supply. An agricultural drought occurs when soil moisture and rainfall areinadequate during the growing season to support a healthy crop growth to maturitycausing extreme crop stress and a drastic fall in yields. While meteorologicaldroughts result from precipitation deficiencies; agricultural droughts are largely theresult of soil moisture deficiencies. A plant’s demand for water is dependent on theprevailing weather conditions, its genetic characteristics, its stage of growth and thephysical and biological properties of the soil. An operational definition ofagricultural drought should take into account the susceptibility of crops to extrememeteorological conditions at different stages of their development. For example,deficient subsoil moisture during an early growth phase will have little impact onfinal crop yield if topsoil moisture is sufficient to meet early growth requirements.However, if the deficiency of subsoil moisture continues, a substantial loss in yieldmay result.

Finally, socio-economic drought refers to drought attributed to the joint effects ofnatural and societal factors. This type of drought is associated with the supply anddemand of some economic goods. It occurs as an interaction between agriculturalactivity (i.e. demand) and natural events (i.e. supply) resulting in inadequate watervolume or quality for plant and/or animal needs. The supply of some economic goods(e.g. water, hay, electric power, etc.) is weather dependent. In most instances thedemand for goods in general is increasing as a result of increasing population and/orper capita consumption. Therefore socio-economic drought could be defined asoccurring when demand exceeds supply as a result of weather related shortfalls(Sandford, 1979). This concept of drought supports the strong symbiosis that existsbetween drought and human activities. This type of drought is related to factors such asthe distribution of plants, animal and human populations, life style, land use, etc.

2.1.3DROUGHT CONCEPTS, DEFINITIONS

AND QUANTIFICATIONS


Most drought monitoring systems are based largely on meteorological data. Suchsystems are valuable as a first stage in drought assessment. Meteorological data canbe used in conjunction with other data and information to estimate the probableimpact of a drought. Identification of drought of any category and its extent needsdifferent types of data depending upon the purpose for which the drought is beingdefined. The most common data for all types of drought assessment is rainfall.Hydrological drought monitoring requires rainfall and evaporation data and information on the water holding capacity of the soils in the catchment areas of thewater bodies. For agricultural drought assessment data on soil type, its texture,water holding capacity, the slope of the surface, soil bulk density, cultivar characteristics, irrigation and crop management are needed.

Among developing countries, India now has systematic data, covering morethan a 100 years, on important weather factors such as rainfall, temperature,atmospheric pressure, winds, etc. for a number of stations. For information relatingto earlier centuries, one has to fall back on cursory records and such indirectinformation on weather patterns as is provided by growth rings of old trees, i.e.dendrochronology. As far as agricultural drought is concerned, the waterrequirements of crops vary considerably between different crops and betweendifferent stages of the same crop. A detailed study of crop water requirements ispresented by Doorenbos and Pruitt (1977) and Yao (1981). The impact of droughtswill not be well understood unless the key phases of crop development especiallysusceptible to adverse weather conditions are considered. Data on crop type anddevelopmental stage should also be collected and analyzed for calculations of cropspecific evapotranspiration. Criteria and thresholds for the onset and cessation ofdrought conditions must be developed, integrating meteorological and crop data forspecific regions. Unfortunately, such data are greatly lacking in most countries. InIndia, reliable data is now available from nearly 200 agrometeorologicalobservatories and 620 observatories recording evaporation. Data on theevapotranspiration loss of crops are also available from 39 stations.

In arid zones, much of the rainfall is lost by surface run-off or evaporation.Countries such as Tunisia, Jordan and Syria have reported successful utilization ofsurface run-off water for irrigation. Unfortunately in most cases, there is no accuratequantification of surface run-off.

Strong emphasis must be placed on the reliability and timeliness of data. Datamust be collected at an adequate spatial density to properly represent droughtconditions and it must be of sufficiently high quality to allow accurate assessment.Information on the onset, severity, spatial extent and the probable impacts of thedrought are not always disseminated to users in a real-time mode. Informationshould reach users in time to be incorporated in the decision-making process. It isimperative that the timing of critical decisions by primary users be taken intoaccount. It is also essential that data and information delivery systems be developedin concert with user requirements and that educational programmes be madeavailable to primary users to train them in product application. Lines ofcommunication must be established with all primary users and they must be openat all times.

Monitoring, detection and reporting systems are generally more effective if theyare built on several independent data collection networks. Three main types of datacollection networks exist for this purpose:

(a) Networks of surface-based instruments, including both low (e.g. manual weatherobservation networks) and high technology (e.g. automated weather observationnetworks) types;

(b) Satellite imagery; and(c) On-site inspections.

Ironically, the so-called low technology methods of communication are not afeasible option for transmitting data and information in many developing countriesbecause of poor basic infrastructure. Automated data collection or high technologysystems that can transfer many kinds of data from meteorological and agronomicsensors through surface-based or satellite linkages are becoming more affordable. Nodata collection system is however complete unless it includes an efficient and

2.1.4DATA AVAILABILITY

CHAPTER 2 — AGROMETEOROLOGICAL ASPECTS OF DROUGHT AND DESERTIFICATION 9

effective means of communication from the observation point to the processing oranalyzing point. Adequate quality control, regular instrument maintenance, efficientprocedures, and communication channels for transmitting advice and warnings tousers are also essential. Maintaining or enhancing existing networks and/orestablishing new data collection networks is generally costly, but this is essential toensure a dependable monitoring system. An inventory may identify areas of datadeficiency (quantity or quality) that must be addressed. Needless to say,meteorological data represent an important part of any drought monitoring system.

Conventional surface observation stations within National Meteorologicalservices provide essential benchmark data and time series necessary for improvedmonitoring of the climate and hydrologic system. Currently, many observationalnetworks, especially in developing countries, do not provide sufficient informationfor some user applications. Reporting networks also need to be upgraded by addingautomated stations to provide more timely reporting of data and/or data gatheringfrom remote locations. All these have been defined by Wilhite (1990) as the data-information continuum (Figure 2.1). Incidentally, automated networks have beenoperating on a routine basis in many developed and developing nations. In thisconnection, the use of Advanced Very High Resolution Radiometer (AVHRR)digital data from the GOES satellite, operated by NOAA (Tucker and Goward,1987) should be considered seriously. These data are transmitted by the satellite infive discrete bands of the electromagnetic spectrum, two of which are useful for landresource investigations. These data can be used to depict changes in thephotosynthetic activity of vegetation and thus are useful in the early detection ofthe onset and spread of drought conditions. These data are used routinely as part ofthe Famine Early Warning System (FEWS) in Africa and the National AgriculturalDrought Surveillance System in India (Thiruvengadacheri, 1991); many othernations also use the data.

Drought is a regional manifestation of general climatic fluctuations associated withpersistent large-scale aberrations of the atmospheric circulation. Meteorologistsusually explain drought in a given region in terms of the abnormal atmospheric

2.1.5CAUSES OF DROUGHT


DETERMINE PRIMARY USERS AND USERS’ NEEDS

EDUCATE USERS ON THE USE OF INFORMATION

DATA COLLECTION

DATA QUALITY CONTROL

DATA ANALYSIS

DATA INTERPRETATION

PRODUCT REVISION

DEVELOPMENT OF PRODUCT

DEVELOPMENT OF A DELIVERY SYSTEM

DISSEMINATION OF INFORMATION

PRODUCT EVALUATION

Figure 2.1The data-information

continuum for thedevelopment of

weather/climate data-basedproducts (Wilhite, 1990)

circulation patterns which favour subsidence over the region. Although ultimatecauses for climatic fluctuations and variations have not been identified with certainty, factors may include changes in the composition of the atmosphere, large-scale volcanic activity and episodes in sea-surface temperature in the equatorialPacific (called El Niño), etc. Identification of the consistent El Niño SouthernOscillation (ENSO) precipitation relationship provides the clearest indicator thatseasonal meteorological drought in quite extensive areas of the globe may be predictable. The link with ENSO can be formalized by calculating precipitationprobability distributions conditional on the state of ENSO (Ropelewski, 1995).

The importance of precipitation to agriculture is obvious, being the principal sourceof moisture required for crops and livestock. Timing and frequency of precipitationare extremely important. Short-term episodes of dryness may be of littleconsequence to crops at a particular phase of their growth cycle, while a similaroccurrence during a highly sensitive phase may ultimately ruin the crop anddrastically reduce the yield potential. Identifying the onset of drought shouldconsider not only the normal climatic regime but also regional agricultural zones.The commencement of drought need not coincide with the beginning of a dry spell,since as far as agricultural drought is considered, the crops may survive for some timeon stored moisture. The depletion of water levels in dams and rivers, whichconnotes meteorological drought, is felt only after the level drops substantially. Thecessation of a drought episode is also difficult to determine because brief intermittentinterruptions of the drought event may be of little significance to agriculture(WMO, 1992). A heavy rainfall event following a long dry episode may producemore run-off than moisture penetration into the parched soil and hence, thoughuseful for terminating hydrological drought, may be of little consequence toagriculture.

The onset of drought may be hastened by anomalies in high temperatures, highvapour deficits and strong winds. These should be factored into the drought analysis.Another distinguishing feature of drought is its duration. Droughts usually require aminimum of two to three months to become established and usually last for morethan a season, but sometimes for several years or even decades. The magnitude ofdrought impacts is closely related to the timing of the onset of the drought, itsintensity and the duration of the event. Droughts also differ in terms of their spatialcharacteristics. They can occur over areas of a few hundred square kilometres butalmost invariably intensities are not uniformly severe and duration is relativelyshort. On the other hand, continental drought may extend over vast areas coveringhundreds of thousands of square kilometres. In large countries, such as Brazil, China,India, the United States and Australia, drought rarely affects the entire country.

Drought is a phenomenon that can be realized only after it has occurred. Theimportance of drought lies in its impact on agriculture and related socio-economicfactors.

In agriculture, severe drought is the phenomenon where soil moisture is insufficientto meet normal plant development and growth requirements. It is a complexphenomenon affected by a number of factors namely:

(a) Meteorological factors, i.e. precipitation and its intensity, air and soil temperature,solar radiation and sunshine duration and wind speed;

(b) Agricultural factors, i.e. the type of crop, stage of plant development and methodof cultivation;

(c) Natural environment factors, i.e. soil, hydrology and drainage of the soil, topographyand land forms; and

(d) Irrigation and anthropogenic factors associated with land use practices, notablydeforestation and over grazing which tend to modify the surface reflectivity (albedo),surface roughness and moisture convergence. These affect the feedback on moisture recycling mechanisms leading to reducedevaporation and hence available atmospheric moisture required for cloud formationand hence precipitation (Gbeckor-Kove, 1995).

2.1.7.1Drought and its agricultural impact

2.1.7IMPACT OF DROUGHT

2.1.6SPATIAL AND TEMPORAL ASPECTS OF

DROUGHT


Lack of water in the soil results in less water being absorbed by the roots of plantwhile at the same time more is being transpired by the leaves. As a consequence,the water budget within the body of a plant becomes unbalanced leading to waterdeficit. Water deficiencies during the growing season often result in stunted ordistorted development and reduced crop yields. When hardened plants are subjectedto drought, their protoplasm shows a lower viscosity and higher permeability towater than that of similar but non-hardened plants. Higher rates of photosynthesis,lower rates of respiration and a higher root/shoot ratio characterize hardened plantscontributing to large yields. The reverse is the case for non-hardened plants. Wheneffective moisture in the soil decreases to a certain level, plant roots are hinderedfrom absorbing moisture and the plant begins to wilt. In due course permanent wiltbrings severe damage to plant.

Drought conditions also lead to:(a) Shortage of food production due to crop failure;(b) Shortage of fodder and drinking water for cattle, migration of livestock population

and even decrease in animal population;(c) Shortage of draught power for agricultural operations during the subsequent period

as a result of the reduced animal population; and(d) Deforestation because of increased fuelwood needs due to the non-availability of

agricultural wastes and crop residues.Droughts have an immediate effect on the recharge of soil moisture resulting

in reductions of streamflow reservoir levels and irrigation potential and even theavailability of drinking water from wells. In regions dependent on groundwater forirrigation, poor farmers are affected because their wells are shallow. Rich farmershave deeper wells and can afford higher pumping costs. Thus, drought increasesdisparities. When drought occurs consecutively for two years, hydropowergeneration is also adversely affected (Wilhite, 1993). Indirect effects of drought onagriculture-related activities:In the event of a prolonged drought, farmers, either alone or with the entire family,may abandon their land in search of work and food in nearby cities. Fewer andweaker family members remain to till the land, affecting the area under cultivation(McCann, 1986). The acreage planted to food crops is also affected by land quality. Due to the uncertainty of rains during the drought, farmers sometimesmake several attempts at sowing of seeds leading to a drastic reduction in seedreserves, which in due course are neither sufficient for planting nor forconsumption. The farmer is then obliged to borrow, offering labour or perhaps aportion of the future harvest as payment for the loan.Drought not only exposes and accelerates existing land quality problems, it also initiates new ones. The cultivation of lands subject to a high degree of rainfall variability makes them extremely susceptible to wind erosion (and desertification)during prolonged drought episodes, as the bare soil lacks the dense vegetative covernecessary to minimize the effects of aeolian processes. As the fertility of the landand crop yields decline farmers search for new land to cultivate. Farmers are sometimes forced to cultivate lands considered marginal from the viewpoint of soilquality, terrain slope and rainfall (Glantz, et al., 1986). These newly cultivatedlands are high risk areas in the long run for rainfed agriculture.Governments usually give certain agricultural commodities favoured treatment forexport to earn foreign exchange. Even cash crops grown in relatively fertile andbetter watered areas are not immune from the secondary effects of prolongeddroughts and fail to attain their full potential. Generally, foreign exchange earnedfrom cash crops is not used for agricultural development; instead it is often divertedto support non-development related programmes. Drought affects labour supply inthe agricultural sector; men often leave their villages in search of income-generatingwork. This robs the rural areas of the much-needed labour supply for agriculturalactivities and leads to a fall in agricultural production (Glantz, 1987).

Other socio-economic implications of droughts include a rise in the prices of essentialcommodities, import of food grains, distress sale of cattle, rural unemployment,health hazards, depletion of farmers assets, the spread of refugees, social instability

2.1.7.2Socio-economic impacts

Decreased export potential

Land degradation

Area under cultivation


and deaths due to malnutrition and hunger. Desertification and a reduction of waterresources are also indirect effects of drought. Drought impacts are long lasting, attimes lingering for many years. Human and social factors aggravate the effect ofdrought as it takes several years for small and marginal farmers in drought areas torecoup the losses. Wilhite and Wood (1994) presented a comprehensive list of theimpacts of drought classified as economic, environmental and social.

While we have focused on the negative impacts of drought, it does also have somebeneficial aspects. Generally, the beneficial aspects are less emphasized and mayhave smaller economic and social significance than the negative aspects. Moderatedrought in the post flowering maturity stage of sugar cane, for instance, helps toincrease the sucrose content.

Other beneficial aspects include mosquito reduction, reduced cost for snowremoval in snowfall regions and other related activities, emergency waterconservation leading to the permanent and efficient establishment of water savinguse patterns, etc. At the secondary level, drought may reduce populationimmigration to areas especially prone to droughts.

Droughts may help control overproduction in agriculture and other sectors, thuscontributing to more stable prices and survival of farming communities which maybe threatened by overproduction.

Drought is a phenomenon associated with water scarcity. The period during whichthe scarcity is likely to be experienced, the extent of the scarcity and also theareas/regions that are likely to be affected by drought have to be known in advance.Such information can be utilized to make an assessment of the impact of droughtand identify suitable mitigation measures. This problem can be investigated by subjecting the drought data, particularly rainfall, to statistical techniques to explorethe possibilities of any systematic pattern – trend, persistence or cycle. Kogan (1997)brings out the importance of the antecedent precipitation to some drought impactsas shown by satellite monitoring. However, forecasting of meteorological drought isnot yet operationally possible, although some statistical and dynamical forecastshave been prepared experimentally for the Sahel, north-east Brazil, the UnitedStates and Japan (Barnett, et al., 1993; Cane, et al., 1994).

The information on the probable occurrence of droughts of various intensitiescan be used for evolving land use systems and appropriate management practicesthat can minimize the impact of drought, in the event of its occurrence. Certainconcepts and related research results that may be useful in moderating the effect ofsoil moisture stress during the occurrence of drought have been described in detail byseveral workers (Katyal, et al., 1992; Hough, et al., 1996). Information on the criticaldates beyond which the sowing of traditional crops/varieties is likely to adverselyaffect crop yields can facilitate the evolution of contingency crop plans under lowrainfall conditions. Finding adequate fodder for cattle during drought years is aserious problem and mobilization of fodder from elsewhere involves enormoustransportation costs. Therefore, fodder security has to be at the top of the agenda indrought prone areas. The generation of raw materials as inputs for agro-based ruralindustries will provide additional employment opportunities for the people living indrought prone areas.

A drought detection, monitoring and early warning system must provide reliableand timely information to users. Through such a system, by using real-time information about the onset of drought conditions, it should be possible to reducethe adverse effects of drought.

Such information could facilitate:(a) Decisions at the time of sowing about the choice of crops/varieties;(b) The adoption of management practices related to soil moisture conservation, fertilizer

application and thinning of plant populations during the crop growing period.The onset of drought is gradual and its intensity develops slowly over a prolonged

period. These features of drought have made monitoring drought conditions adifficult task. In general, two types of monitoring systems exist: surface observation

2.1.9DROUGHT DETECTION, MONITORING

AND EARLY WARNING

2.1.8FORECASTING DROUGHT

2.1.7.2Beneficial aspects of drought


and satellite/remote sensing networks. Using both methods, a three-dimensionalmonitoring system can be established, characterized by combinations of ground-basedobservation and space-based and mobile observation facilities. These networks arecapable of monitoring all the critical elements associated with drought, includingprecipitation, surface water, soil moisture, crop water requirement, groundwater,irrigation and drainage. In addition, drought indices can be calculated using datacollected from these networks. It is vital to transmit this information in a timelyfashion via suitable telecommunications systems. A properly organized deliverysystem is essential for the monitoring and early warning system to be effective as adisaster mitigation tool. On the basis of the information generated through thissystem, emergency activities can be better coordinated. This information, used incombination with weather/drought forecasts, can provide decision-makers with theinformation needed for disaster management and also provide a scientific rationale forgovernment and other sectors to make short-, medium- and long-range decisions. Indrought monitoring, it is essential to establish appropriate and reliable droughtindices for different user groups.

A drought index calculated from known values of selected parameters enables thedescription of drought to be expressed quantitatively. There are two primary usesfor an index of drought, namely for evaluating the drought hazard over an area andin assessment of the current extent and severity of drought over a region. Themajority of indices reflect the meteorological drought but not the agricultural one.These indices, as such, cannot serve the purpose of drought problems linked to cropproduction. The problem of agricultural drought pertains to the physical and bio-logical characteristics of plants and their interaction with the environment.Hydrological drought differs from meteorological drought in that the streamflowrate, water reservoir supplies and groundwater levels are affected by longer durations of unseasonable dryness.Most meteorological drought indices use deviation of the seasonal or monthly valueof various weather elements from a central value to quantify drought severity. Witha view to assessing deficiency in monsoon rainfall in India, an index termed theMonsoon Deficiency Index was developed by Mooley and Parthsarthy (1982). Thisindex was obtained by expressing the area of the country receiving 80 per cent ofthe normal seasonal rainfall as a fraction of the total area of the country. Mooleyand Parthsarthy (1983) proposed another criterion based on rainfall expressed as astandard deviate, Yi given by:

Yi = (Xi – X–)/σ

where Xi is the rainfall of ith year; X– the normal rainfall; and σ the standarddeviation. They consider drought to have occurred when Yi < –1.28, the value 1.28being 10 per cent value of the Gaussian distribution.

Gibbs (1964) has shown that the mean and standard deviation of the squareroots can give an adequate description of the entire rainfall distribution. Thepercentile and decile methods are also often used. The fifth decile or the median isthe amount that was exceeded on 50 per cent of occasions. The first decile range(i.e. the range of values below the first decile) implies abnormally dry conditions,while the tenth decile range (i.e. above the ninth decile) implies very wetconditions.

The most elaborate and perhaps the most satisfactory indices are the DroughtSeverity Index derived by Palmer (1965) for Colorado and his Crop Moisture Index(1968); the former reflects meteorological drought and the latter gives an idea ofthe severity of agricultural drought.

The Palmer Drought Severity Index (PDSI) relates drought severity to theaccumulated weighted differences between actual precipitation and the precipitationrequirements of evapotranspiration. The PDSI is based on the concept of anhydraulic accounting system and is actually used to evaluate prolonged periods ofabnormally wet or abnormally dry weather. McKee, et al. (1993) developed theStandardized Precipitation Index (SPI) as an alternative to Palmer’s index.

Meteorological drought indices

2.1.9.1Drought indices


Historical data are used to compute the probability distribution of the monthly andseasonal observed precipitation totals and then the probabilities are normalized.This methodology also allows expression of droughts (and wet spells) in terms ofprecipitation deficit, percent of normal and probability of non-exceedance. WMO(1975) gave a number of drought indices based on rainfall and temperature formeteorological drought.Agricultural drought indices are mainly developed from rainfall, evaporation, evapotranspiration, water status of soil at different stages of crop development, etc.and have been used to quantify the severity of agricultural drought. Agriculturaldrought begins when the vegetation cannot extract water from the soil rapidlyenough to replace the moisture loss by respiration. It persists when there is no continued replenishment of the water in the soil. Most of the agricultural droughtindices are used to express the degree to which the agricultural system has beenaffected by water deficit.

There are many indicators that a plant is suffering from physiological drought,which is related to agricultural drought, such as wilt, leaf colour change and loss ofrigidity, leaf shedding, change in orientation, growth retardation, leaf and stemelongation etc. Considering the complex physical processes involved in the responseof the agricultural systems to water deficit, it may be very difficult to quantify theeffects of agricultural droughts accurately. Agricultural drough indices have thereforebeen derived from a variety of simple parameters to more complex functionsinvolving a combination of soil moisture, crop parameters and many other factors. Avery useful method of finding the severity of an agricultural drought – the CropMoisture Index (CMI) – was devised by Palmer (1968) by modifying the PDSI. TheCMI defines drought in terms of the magnitude of the computed abnormalevapotranspiration deficit.

Dyer and Baier (1979) developed an index to approximate the drying patternsof various soil types. This replaced an earlier approach of selecting a table ofcoefficients for each drying curve. This technique was used in the “Versatile SoilMoisture Budget” model (Baier and Robertson, 1966). Jackson (1982) presented atheoretical method for calculating a Crop Water Stress Index (CWSI), requiringestimates of canopy temperature, air temperature, vapour pressure deficit, netradiation and wind speed. The CWSI was found to hold promise for improving theevaluation of plant water stress. The use of canopy temperature as a plant’s droughtindicator and stress is used by Idso, et al. (1980) to calculate the Stress Degree Day(SDD) index. The cumulative value is related to final yields.

Other indices have considered the major agrometeorological factors which areseverely affected by moisture deficits. Such parameters include the biologicalcondition of plants, soil type, nutrient constraints, stages of plant development, finalcrop yield and several other agrometeorological factors. Many of these methods havebeen discussed by WMO (1971, 1975a, 1975b, 1983, etc.).The frequency and severity of hydrological drought is often defined on the basis ofits influence on the river basin. Hydrological droughts are often characterized bylow streamflow, low precipitation, a fall in the levels of lakes, wells and reservoirs,depletion of soil moisture, lowering of groundwater tables, changes in run-off patterns, evaporation rates, etc. Changes in these parameters together with otherhydrological factors have been used to quantify the severity of hydrologicaldroughts (Chow, 1964; WMO, 1969, 1983a). Water balance indices can also bederived for the hydrological system. Although the PDSI is sometimes used as anindicator of hydrologic drought, other definitions have been formulated whichbetter serve the needs of hydrologists. For example, a definition of hydrologicaldrought was developed in Colorado in 1981.

To assess drought conditions in high elevation river basins that are dependenton snow melt as their main source of water supply (Shafer and Dezman, 1982), theSurface Water Supply Index (SWSI) was intended to be complementary to thePDSI, with the latter applying mainly to non-irrigated areas independent ofmountain water supplies. The SWSI integrates historical data with current reservoirfigures and precipitation at high elevation into a single streamflow.

Hydrological drought indices

Agricultural drought indices


The ability to use satellite data to detect drought conditions is based on the spectral manifestations of reduced photosynthetic capacity which is associated withprecipitation shortfalls. By analyzing several years of satellite data, we can makecomparisons between years in terms of estimated photosynthetic capacity. Lowerthan average conditions provide the means to substantiate the occurrence ofdrought. Monitoring by remote sensing techniques is most appropriate for detecting the status of crop growth, soil moisture, evapotranspiration and precipitation. Using a surface observation station network and remote sensingtechniques, development and spread of drought conditions can be monitored in aroutine and cost effective manner. Since 1979, the Assessment and InformationService Centre (AISC) of the NOAA/NESDIS has been providing drought earlywarning alerts and climate impact assessments to national and international agencies that require such information for disaster preparedness and agriculturalassessment (Sakamoto and Steyaert, 1987). The Drought Early Warning programme of AISC is an operational programme that includes assessment modelling, assessment reporting and development of technical assistance in lessdeveloped countries.

Drought does not descend all of a sudden. It results from a set of weather sequencesthat requires an extended period to develop. Thus it takes a long period for adrought situation to begin, expand and decay – allowing time to adopt contingencyplans to reduce the adverse effects of drought. There are two distinct phases inwhich the application of weather and climate knowledge can reduce the impact ofdrought on communities. The first is long-term planning in which strategies can bedevised, and precautions taken to reduce impact. The second phase is the action tobe taken to reduce the adverse effects during the onset of the event.

In arid, semi-arid and marginal areas with a probability of drought incidence,it is important for those responsible for land use planning to seek expertclimatological advice regarding rainfall expectations. Drought is the result of theinteraction of a human pattern of land use and rainfall regimes. In these regions adetailed examination of rainfall records is essential. In this regard, the developmentof methods of predicting, many weeks or months in advance, the occurrence ofrainfall deserves high priority.

Since technological inputs quickly reach an optimum level, more emphasisshould be placed on drought management policies, especially in dryland farmingareas. Agricultural planning and practices need to be worked out with considerationto the overall water requirements within an individual agroclimatic zone. Cropswhich need a short duration to mature and require relatively little water need to beencouraged in drought prone areas. Irrigation, through canals and groundwaterresources, need to be monitored with optimum utilization avoiding soil salinity andexcessive evaporation loss. A food reserve is needed to meet the emergencyrequirements of up to two consecutive droughts. A variety of policy decisions onfarming, human migration, population dynamics, livestock survival, ecology, etc.must be formulated (Das, 1999).

Sustainable strategies must be developed to alleviate the impact of drought oncrop productivity. In areas of recurring drought, one of the best strategies for alleviatingdrought is varietal manipulation. By adopting varieties that are drought-resistant atdifferent growth stages, the effects of drought can be avoided or minimized.

If drought occurs during the middle of a growing season, corrective measurescan be adopted; these can include reducing the plant population, fertilization andweed management. In high rainfall areas where there are a series of wet and dryspells, rainfall can be harvested in either farm ponds or in village tanks and can berecycled as lifesaving irrigation during a prolonged dry spell. The remaining watercan also be used to provide irrigation for a second crop with a lower waterrequirement, such as chickpea.

However, no one strategy can be adopted universally. In fact, all such strategiesare location, time, crop, crop stage and (to some extent) socio-economic conditionspecific. Developing such strategies for each specific factor can help make agriculturesustainable.

2.1.10ADAPTATION AND ADJUSTMENTS

TO DROUGHT

2.1.9.2Monitoring by satellite remote

sensing


Once drought has set in, individual farmers, in an effort to adapt and adjust,often: (i) reduce consumption; (ii) postpone social arrangements such as marriages;(iii) migrate to better areas with livestock or sell livestock; (iv) take loans; and (v)sell assets such as gold ornaments (Venkateswarlu, 1987).

Only very few farmers are able to store food grains and fodder to tide them overthe crisis during the years of drought. Some strategies also believed to alleviatedrought conditions include: (i) groundwater exploitation; (i) soil and waterconservation and management; (iii) intercropping; (iv) introduction of alternativecrops/varieties; (v) afforestation; and (vi) the creation of storage facilities for foodand fodder by constructing of rural godowns.

Drought is a recurring phenomenon and its occurrence cannot be avoided.However, its impact can be minimized through the application of science and technology in developing suitable drought management plans (Das, 1995). Usuallywithin a drought affected region, while some areas are devastated there are alwaysothers which remain unaffected. It is important to develop infrastructure for mitigating drought. Drought planning and water crisis management need to beproactive – the overall policy, legislation and specific mitigation strategies shouldbe in place well before a drought or water crisis affects the regular supply of thecountry’s water resources. Ad hoc crisis management is inferior and more costlythan implementing a pre-planned crisis policy. Hastily prepared assessment andresponse procedures may lead to an ineffective, poorly coordinated and untimelyresponse (Wilhite and Easterling, 1987; Bruins, 1993).

The analysis of drought risk can be investigated on the basis of meteorological,paleoclimatic and historical data on climatic variations (Bruins, 1994; Issar, et al.,1995). The timescale in which severe droughts recur may exceed the averagehuman lifespan. Perception of both risk and impact may therefore disappear frompublic memory – this can have a bearing on planning and policy by the authorities.Drought definitions need to be precise, regional and even specifically targeted atselected economic activities to be useful for government policy and proactiveplanning.

Assessments have to be made of the impact of drought on the various waterresources, economic sectors, towns, villages and the environment. Respectivevulnerability of different sections of the environment, economic activities and socialgroups needs to be studied at different levels – local, provincial, national andregional (Bruins and Lithwick, 1998). Drought scenarios have to be drawn up onthe basis of available information, including frequency and severity, if applicable.Finally, proactive drought contingency planning needs to be developed. Wilhite(1993) outlined in considerable detail a generic process with meteorological stepsthat may be adopted by governments (Wilhite, 1986; Wilhite and Hayes, 1998) todevelop comprehensive drought planning and management. Proactive planningneeds to be executed through interactive management (Bruins and Lithwick, 1998)to ensure planning for future drought situations is as comprehensive as possible.Actual realization of the proactive plans in times of drought needs to be adjustedand updated through interactive management.

The major issues that need to be addressed in this connection are:(a) Research efforts must deployed in making a reliable assessment of the likely impact

of the drought;(b) Availability of resources such as credit, fertilizers, pesticides and power for

increasing production;(c) Organization of buffer stocks of food grains and fodder to cope up with the likely or

anticipated shortages;(d) Coordination of the activities of various agencies in implementing drought

management plans;(e) Education of the population on the various mechanisms available to cope up with

the drought situation;(f) The enhancement of local employment opportunities to reduce the percentage of

the population dependent upon agriculture in drought prone areas or their migration to the cities; and

2.1.11DROUGHT MANAGEMENT:

MITIGATION, PREPAREDNESS

AND POLICY


(g) Providing economic relief through the creation of durable assets rather thanextending subsidies.

Coping strategies for responding to and preparing for drought are numerous andrange from the individual or household level to the national level, as described inthis chapter. Government policy responses to drought can be broadly classified intothree types (Parry and Carter, 1987): pre-impact programmes for impact reduction,post-impact government interventions and contingency arrangements orpreparedness plans. Pre-impact government programmes are defined as those thatattempt to mitigate the future effects of drought. Specific drought-related examplesinclude the development of an early warning system, augmentation of watersupplies, demand reduction (such as water conservation programmes) and cropinsurance.

Post-impact government interventions refer to those reactive programmes ortactics implemented by government in response to drought or some other extremeclimatic event. This includes a wide range of reactive emergency measures such aslow-interest loans, transportation subsidies for livestock and livestock feed, provisionof food and water via tankers, drilling wells for irrigation and public water supplies.This reactive crisis management approach has been criticised by scientists,government officials and many relief recipients as inefficient, ineffective anduntimely (Wilhite, 1993). More recently, the provision of emergency relief in timesof drought has also been criticized as being a disincentive to the sustainable use ofnatural resources because it does not promote self-reliance (Bruwer, 1993; White,et al., 1993). In fact, this approach may increase vulnerability to drought as well toother natural hazards.

Contingency arrangements refer to policies and plans that can be useful inpreparing for drought. These are usually developed at national and provincial levels,with linkages to the local level. The ultimate goal of these preparedness plans is toreduce vulnerability to future episodes of drought. Until recently, nations haddevoted little effort to drought preparedness, preferring instead the traditionalreactive or crisis management approach.

Deficiencies of previous drought assessment and response efforts are welldocumented (Wilhite, 1992). They include:

(a) Lack of appropriate climatic indices and early warning systems, as well as a lack oftriggers for initiating specific actions;

(b) Insufficient databases for assessing water shortages and potential impacts; (c) Inadequate tools and methodologies for early estimates of impacts in various

sectors; (d) Insufficient information flow within and between levels of government on drought

severity, impacts and appropriate policy responses; (e) Inappropriate or untimely emergency assistance programmes; (f) Poorly targeted emergency assistance programmes that do not reach vulnerable

population groups and economic sectors; (g) Meagre financial and human resources that are poorly allocated; (h) Lack of emphasis on proactive mitigation programmes aimed at reducing

vulnerability to drought; (i) Institutional deficiencies that inhibit effective emergency response; and(j) Lack of coordination of policies and programmes within (horizontal) and between

(vertical) levels of government.Institutional, political, budgetary and human resource constraints often make

drought planning difficult (Wilhite and Easterling, 1987). One major constrain thatexists worldwide is a lack of understanding of drought by politicians, policy makers,technical staff and the general public. The recent efforts to combat drought throughpolicies formulated by the government agencies include: (i) crop weather watchgroups at national and state levels; (ii) food security through buffer stocks; (iii) priority in the most seriously affected areas through the “food forwork”/National Rural Employment Project and other programmes; (iv) high priorityof food production in the most favourable/irrigated areas as compensatoryprogrammes; (v) optimum input use; (vi) the construction of rural godowns to avoiddistress sales; and (vii) crop insurance schemes.


The approaches and types of programmes in response to drought, whether longterm or short term should be constantly scrutinized to find ways of improving themand each drought phenomenon must be followed by a review of the functioning ofthe organization and of the public response.

Although in one sense drought is essentially a physical phenomenon as describedabove, in another sense it can be considered essentially a socio-economic phenomenon. Physical drought encompasses various hydrometeorological characteristics, while socio-economic drought evaluates the effects of the samevariables on the general well-being of the society.

Drought characteristics and impacts have both long-term and short-termeffects. Short-term effects are generally well known. Dryland agriculture is often thefirst to experience the direct effects of drought, usually in the form of reduced yieldsand dust blowing winds. When soil moisture levels drop during drought situations,transpiration and plant growth decrease, the latter being more pronounced than theformer. Irrigated agriculture also suffers directly. When household water provisionis disrupted, while economically less important, the drought is brought home toeveryone and, in extreme cases, can create severe problems in health, nutrition andsanitation.

The long-term effects of drought are more subtle and difficult to assess, but it isreasonable to assert that their magnitudes could exceed short-term effects. Very fewof the expected long-term drought effects have been studied in sufficient detail toassess the magnitude of economic losses and social inconvenience andmaladjustment. Likewise, there is insufficient knowledge for developing andassessing cost-effective means of mitigating long-term effects.

At regional, and especially at national levels, the policy issues andcorresponding drought strategies acquire different dimensions. Here the imperativesare not simply for pooling physical requirements in multipurpose, basinwide,integrated water systems, but rather for actions which facilitate provision of credit,technological expertise, research funding, etc. Other devices include the provisionof drought insurance, loans, grants, etc.

The use of drought resistant crops and vegetation can be an effective way tocope with restricted or limited water availability during droughts. Agronomicresearch in this area requires proper identification of drought resistant plant varietiesand various possible conditions as defined by such variables as rainfall distribution,soil type and growing season. This knowledge, in addition to proper quantificationand evaluation of drought resistance, should accelerate development of plant hybridsresistant to droughts. Continuing research on the response of crop varieties tolimited or scarce water supplies is needed. Where water is available, irrigation andconservation practices may provide effective ways of agricultural adjustment todrought. In these cases, aspects such as timing of irrigation, management of landand crop use, etc. need to be studied.

2.2 DESERTIFICATION

Desertification is a worldwide phenomenon affecting all continents but it mostlyaffects the arid and semi-arid zones of the world. In recent years this menace seemsto be accelerating most rapidly in developing countries. These countries are experiencing a phenomenal increase in population largely because of their traditional high fertility rate and a reduced mortality rate, made possible by modernhealthcare facilities. This increase in turn imposes a corresponding rise in livestockor in expansion of arable farming on lands which are only marginally suited becauseof adverse soil or climatic factors. Desertification is the spatial extension of desert-like conditions resulting from human impact on the ecosystems of semi-aridregions. It takes place mainly in desert boundary regions and involves a complexphysical geographical processes disturbing the natural ecological equilibrium.Factors which disrupt the ecological equilibrium include the sparse vegetation andthe surface water balance.

2.2.1INTRODUCTION

2.1.12SUMMARY AND CONCLUSION


Desertification results from the over-exploitation of arid and semi-arid lands.There is a widespread, but largely erroneous, belief that desert expansion is caused bydrought. This belief deflects attention away from human responsibility forenvironmental deterioration. There have always been deserts, and there have alwaysbeen droughts, but only in comparatively recent times have normal climaticfluctuations led to starvation and death on so massive a scale as we have witnessedduring the Sahel droughts of 1968–73 and 1982–85 (Dregne, 1983).

Desertification can only be halted by encouraging multiple land use, exploitingthe natural diversity of the desert biome and, at the same time, exercising restraintin the size and scale of developmental projects in the fragile arid ecosystems. For itto be effective, however, the cooperation of local inhabitants – frequently neglectedin the past – is absolutely essential. Not only must they be given alternative foodand employment if their flocks of sheep and goats are to be reduced, but they alsoneed to be provided with fuel oil if they are not to cut down trees for charcoal tocook with. We cannot expect the poor of developing countries to deny themselvesthe sustenance of life for the benefit of mankind in general – even though desert-dwellers are renowned for their generosity and hospitality.

Desertification is a degradation of an ecosystem resulting in a desert-like environment in arid, semi-arid and some sub-humid zones. According toFAO/UNEP, desertification is a process involving all forms of degradation (naturalor human-induced processes disturbing the equilibrium of soil, vegetation, air andwater) of land vulnerable to severe aridity, leading to the reduction or destructionof the biological potential of the land, deterioration of living standards and intensification of desert-like conditions (WMO, 1985). It is perceived as a packageof processes which brings about certain basic changes in a particular ecosystem andconverts it from relatively non-desert to desert terrain. This involves an interplayof climate, edaphic and biotic factors.

2.2.2DEFINITION OF DESERTIFICATION


Figure 2.2The occurrence of drought (January 1982 to August 1983)

Recently the United Nations Conference on Environment and Development(UNCED) adopted the definition of desertification as land degradation in arid, semi-arid and dry sub-humid areas resulting from various factors including climaticvariations and human activities. It is a widespread but discrete process of landdegradation in space through the drylands. The process of desertification ismanifested by an increase of aridity, invasion of blown sands, loss of moisture,accumulation of salt in soil, decline of soil fertility, reduction of vegetative coverage,change of species composition and enlargement of the extent of sandy areas.

Desertified lands are mainly located in arid and semi-arid zones with some of themappearing even in sub-humid zones (Figure 2.3). Desertification means not only theencroachment of existing deserts but also the appearance of desert-like landscapes in originally non-desert areas. The area of land affected or threatened bydesertification is about 39.4 million square kilometres which is 26.3 per cent oftotal land (Hopkins and Jones, 1983). The distribution of these desertified lands areas follows: 36 per cent (about 14.2 million square kilometres) in Asia; 25.4 per cent(about 10.4 million square kilometres) in Africa and 11.8 per cent (about 4.65 million square kilometres) in North and Central America. More than 100 countriesand regions are confronted by this problem. There are about 50–70 million squarekilometres of land likely to be desertified every year, according to one estimate. Atthe end of this century, one third of arable lands will probably be lost, directlythreatening 14 per cent of the human population living in arid and semi-arid lands.Therefore desertification is one of the most serious environmental problems of thisand next centuries. It has rightly been called the “Earth’s cancer”.

In the Sudano-Sahelian region, desertification is likely to continue to advanceunder the combined pressure of growing human and livestock population, inappropriate agricultural policies and possible further political unrest. Although

2.2.4DESERTIFICATION TRENDS

2.2.3DISTRIBUTION OF DESERTIFICATION


Figure 2.3World distribution of desertification

desertification of the rangelands remains the most widespread problem, the mostsevere hazard in terms of potential environmental damage, lost productivity andpopulation affected, lies in rangelands, croplands and adjacent grazing lands nearthe dry limits of cultivation. The countries showing accelerating trends of desertification include Chad, Djibouti, Ethiopia, Mali, Mauritania, Niger, Senegal,Somalia and Sudan (UNEP, 1984). The regional trends of desertification withinland-use categories and major natural resources are shown in Figure 2.4.


Figure 2.4Regional trends of desertification

within land use categories and major natural resources

(UNEP, 1984)

Accelerating desertification

Continuing desertification

Desertification status unchanged

Status improving

Rainfed Irrigated Forest Ground-Region Rangelands croplands lands woodlands water

resources

Sudano-Sahelian Region

Africa South of Sudano-Sahelian Region

Mediterranean Africa

Western Asia

South Asia

USSR in Asia

China and Mongolia

Australia

Mediterranean Europe

South America

Mexico

North America

Figure 2.5Percentage of drylands affected by

desertification

In North Africa, there are records of success in afforestation and dunestabilization, particularly in Algeria and Libya. Desertification, however, hascontinued in rangelands and in rainfed croplands including areas of arboriculture.This is mainly due to increasing population and livestock pressures, and to thelack of appropriate conservation measures, accentuated by drought.

Desertification operates through several processes such as diminution of vegetativecover, induced instability of surface, soil erosion, encroachment by active sanddunes, the salinization of land, etc. Individual processes need not be mutuallyexclusive. For example, depletion of vegetative cover, deterioration of the physicalcondition of the soil and acceleration of wind erosion can occur together, often in amutually sustaining mechanism. A couple of above normal rainfall years in asequence may lead to a greatly increased vegetation cover and productivity, thuscausing a large deviation from a general declining trend. Likewise, substantial spatial variation can be seen in the degree of deterioration within an area otherwiseuniform in its land attributes. Lands in the immediate vicinity of settlements andwatering points are often devoid of usable vegetation. Variability in intensity of useand degree of management in fact is a major factor that complicates the monitoring of the process of desertification.

The meteorological phenomenon of drought, as has been mentioned earlier, is acombination of atmospheric and soil water stresses leading to disturbance of thewater budget of plants, animals and in extreme situations, human beings.Desertification processes manifest themselves differently in different edaphic andclimatic zones and land use systems. In rainfed semi-arid and arid zone agriculture,desertification is primarily evidenced by practical or total loss of vegetation coverin areas where climatic extremes and high land use pressures have reduced thegrowth of vegetation. In general, natural vegetation is very sparse or scanty in desertification prone areas due to insufficient precipitation. But this scanty vegetation is able to protect and to a large extent stabilize the ground surface.However, when, as a result of animal or human intervention, some of this covergets degraded, wind erosion sets in blowing the topsoil away making permanentplant life impossible due to a lack of water reserves in the remaining shallow soil.

The impact of torrential rains over the decreased vegetative cover disperses thefine surface soil aggregates filling up and sealing the pores. The sealed surface reducesthe infiltration rate, increases runoff, causing sheet and gully water erosion,especially in sloping surfaces. This results in much drier soil conditions and a markeddecrease in vegetation and plant productivity. Patches of land so degraded maygradually grow resulting in desertification.

Two major factors are certainly involved in desertification; the periodic stress of climate and human use and abuse of the sensitive, vulnerable dryland ecosystem. Climatic fluctuations, with changes in the temporal and spatial distribution of rainfall, often result in lengthening aridity phases, higher temperatures and stronger winds. Similarly, increasing human pressure on theecosystem may result in the extension of the cultivated area beyond the borderswhere the human-environment equilibrium is disturbed. Such human pressure normally includes the extension of irrigated areas, improper land use practices insemi-arid and sub-humid areas resulting in land degradation through water andwind erosion, overgrazing by livestock, deforestation for firewood and building,bush and forest fires, salinization, alkalinization and waterlogging. As the degradation of land advances, restoration may become more difficult and expensive. At certain stages, feedback mechanisms may come into play to reinforcethe desertification process.

The natural causes of desertification, as well as those induced by people, are ofan extremely complex character and take varying forms, duration and degree ofintensity and development. Desertification is manifested through the degradationof plant and soil cover and depletion of water resources in an area, i.e. the wholebasis for human well being deteriorates.

2.2.6CAUSES OF DESERTIFICATION

2.2.5.1Agrometeorological aspects in the

desertification process

2.2.5PHYSICAL PROCESSES OF

DESERTIFICATION


Arid and semi-arid ecosystems are extremely fragile and can be easily harmed,since historically they developed under very hard xerothermal conditions, andrecover very slowly when these conditions are disturbed. In certain extreme cases,ecological imbalance results in the irreversible destruction of the biological potentialof the land. Some important factors leading to desertification are mentioned below(Oladipo, 1985; Dhir, 1986; Ogallo and Gbeckor-Kove, 1989).

Broadly, natural factors resulting in desertification can be classified as follows:Climatic factors leading to fragility of desert ecosystems are generally associatedwith inherent soil moisture deficit, high intensity of solar radiation, increasingalbedo of the underlying soil surface, low and erratic precipitation, shifting spatialand temporal distribution of rainfall, high air temperature and dryness, high windspeed, etc. These factors cause high climatic stress.

The Earth’s major desert belt lies in sub-tropical and tropical latitudes wherewidespread and persistent subsidence occurs through the influence of atmosphericcirculations. In these regions of mainly subsiding air, rainfall is not sufficient topermit perennial cultivation, but pastoralism and the growing of drought-resistantcrops (such as sorghum and millet) are practised. Obviously, limited choice of crops,low cropping intensity and low yield potential of conventional crops are thecharacteristics of these regions.

The majority of the arid and semi-arid regions of the world suffer from low anderratic precipitation in addition to high inter-annual variability. This unevenlydistributed rainfall in both space and time causes acute moisture shortage and makesthese regions prone to recurrent and sometimes intense drought.

Though vegetation often recovers after a drought, there are occasions when thedrought is so intense and persistent that the vegetation is not able to recover fully,even when the rains return. As a result, the equilibrium of the ecosystem in theseareas is dramatically lost, initiating the process of desertification.

These are changes in the water regime, such as those in ephemeral surface run-off,which can accelerate or alter the erosion process and can cause permanent, seasonal or temporary sources of water to be much reduced or lost.

These factors are related to the nature of the surface of the land and its underlyingstructure that can be influenced by certain climatic factors giving rise to water andwind erosion.

These factors include potential weaknesses in the soil forming processes, low humusor high carbonate content, high calcareousness and salinity, susceptibility to erosion and waterlogging.

Vegetative factors are those which relate to the nature and behaviour of the plantcover. They include the periodic natural reduction of plant density, the plantgrowth and development cycle, low biomass productivity and an increase in xeromorphic and succulent forms.

One of the most widespread and obvious processes of degradation is the decrease ofvegetation cover through overgrazing. Persistent overgrazing brings about a dramatic change in the cover, although it varies according to natural conditionsand soil type. In the early stages the most palatable species are damaged first.

When the number of animals exceeds the carrying capacity of the ecosystem,pastures do not have time to recover. The situation may be exacerbated at times oftoo frequent or too persistent droughts. When pastures do not have time to recoverbetween droughts, there is even more severe overgrazing as animals graze theremaining grasses even closer and browse shrubs and trees ruthlessly anddestructively. The recovery of the soil to its original condition becomes extremelydifficult.It is often lost permanently.

2.2.6.2Human use and abuse of

vulnerable dryland ecosystems

Overgrazing

Vegetative factors

Pedological factors

Geomorphological factors

Hydrological factors

2.2.6.1Natural factors

Climatic factors


Hooves are far more damaging than mouths. Bare stretches around watering pointsare ample testimony to this. The effect can be seen in a radius of 1 to 2 km from awaterhole. Cattle movement on moist soil with sparse vegetation cover causescompaction. This not only causes a reduction in infiltration but is also a hurdle tothe emergence of seedling and the proliferation of root systems.

Arid and semi-arid lands often carry a sizeable representation of trees and shrubs inthe vegetation cover. The open thorn forest of the Asian sub-tropics and the savannah of Africa are good examples of these. The cutting of these woody speciesfor fuel wood, building materials and bush fencing for enclosures and night sheltersfor livestock seriously depletes this cover. This large-scale cutting leads to the disturbance of soil cover, development of sand deflation and wind and water erosion.

Plant cover degradation over large areas also leads to:(a) An increase in albedo leading to a lower level of absorbed solar radiation;(b) An increase in soil temperature and resulting increase in stress on organisms;(c) The loss of fine materials, both mineral (clay and slit) and organic, due to erosion;

and(d) A reduction in water storage capacity.

These four interacting processes represent local climate changes and thedeterioration of surface microclimate, water budget and soil hydrothermal ultimatelylead to the irreversible processes of desertification. A schematic illustration of theclimatic and anthropogenic causative factors in the desertification process is given inFigure 2.4.

Fires over large areas, which are quite common in arid regions of the world, oftenresult in considerable loss of human life, vegetation, crops, livestock and infrastructure. Consequently, fires are sometimes a cause of desertification in aridand semi-arid areas. Intensive wood fires may destroy the soil cover by burning thedebris layer and organic matter in the upper horizon, resulting in the eliminationof soil fauna, and subsequently in water and wind erosion. Any change in the vegetative cover due to fire may also lead to changes in the microclimatic conditions of the ecosystem, which, in turn, have an effect on the intensity of wateror wind erosion which cause the greatest damage to the fertile layer.

Intensification of arable farming and the opening up of new lands to farming whichare either too sandy and highly erodible, or with shallow soils or are located in situations where climatic conditions permit only occasional success with cropping,have accelerated the process of soil degradation. These lands are not able to recovertheir original cover nor do they sustain farming. On the contrary, cultivation of thesemarginal climatic areas causes increased erosion of the top soil unless vegetable stubble is left on the farm to protect the soil (Olderman and Van Lynden, 1997).

More prevalent in arid zones is the incidence of wind erosion, which causes the lossof the most productive top soil layer, spreads sand sheeting and forms shiftingdunes. The sub-soil may also be less fertile and of poor structure and permeabilityand thereby affect the productivity of the land. The problem is serious in Iraq,Afghanistan, Pakistan and India where dunes are showing increasing signs of instability. As with water erosion, the incidence of wind erosion is also the result ofthe interaction of the erodibility of soil and vegetation cover, besides, of course, thewind regime. Large dust storms that carry particles thousands of kilometres are originated in such areas as the Sahara, north and west China, the south-westUnited States, central Australia, central India and the Russian steppes.

Water is the main problem in dry zones and hence the development of irrigation,despite the huge investment, has been looked upon as a major means of increasingfood and fodder supply. Egypt, Iraq, Sudan and Pakistan have established largecanal irrigation projects. This has permitted a manifold increase in productivity.However, poor water management and over-irrigation have created problems ofwaterlogging and/or salinization. Salinity influences the osmotic process and

Salinization and waterlogging

Accelerated soil erosion, sand sheetingand active dune formation

Extension of arable farming tomarginal lands

Wild bush fires

Lopping and cutting of trees andshrubs for fuel and timber

Trampling


induces physiological dryness in the same way as scarcity of rainfall and high evaporation. Saline seeps resulting from the replacement of perennial deep rootedvegetation by shallow-rooted annual crops have also contributed to desertification,especially on land with a hard sub-soil base material. The problem is then aggravated by fallowing after harvest.

Generally, salinization and waterlogging go together since the latter is alsocaused by irrigation. These two factors act together to reduce yield, limit choice ofcrops to cultivate and ultimately lead to the complete loss of irrigated lands.Salinization is also caused by fertilization.

Other agricultural practices, such as the use of tractors, tillage implements,watering of livestock, etc., can cause soil compaction and crusting. These reduceinfiltration and soil permeability, increase water run-off (and hence erosion) andultimately prevent plant growth. In due course the land becomes barren anddesertified.

Whereas river basin irrigated areas are showing a rise in the water table, in areaswhich are groundwater irrigated aquifer depletion is occurring. The Indian aridzone in Gujarat and Rajasthan and pocket irrigation development in the sub-Saharan region and southern Africa are cases in point. The process is alsoaccompanied in many cases by a deterioration in water quality.

For spreading desertification, several processes may operate simultaneously, feedingback into the system, intensifying the degradation of the quality of the resourcebase and the decline of biological productivity.

The speculation that drought may indeed feed on drought to promote long-term desiccation has led to the postulation of some feedback mechanisms (Hare,1983, 1984; Sabadell, 1982). Among the proposed mechanisms is the albedofeedback hypothesis (Charney, 1975) which postulates that the increased albedo ofdamaged surfaces intensifies atmospheric subsidence and leads to the suppression ofconvection and the maintenance of a desert-type climate, with reduced soil moisturetending to diminish the contribution of latent heat to the atmospheric energybudget (Walker and Rowntree, 1977). The increase in albedo, coupled withoxidation or deflatation of organic litter during prolonged drought, may lead topositive feedback effects in the microclimate with the result that “desert may befeeding on desert”.

Other proposed mechanisms include a reduction of the biogenic supply offreezing nuclei produced as vegetation decays, and a reduction in the surfaceradiative heating, convection and precipitation by increased wind blown dust overa degraded region.

Some of these cause-effect feedback mechanisms, especially the albedo feedbackhypothesis, have been extensively tested by general circulation model (GCM)experiments and have been found to provide some evidence that the concept ofpositive feedback between the surface and the atmosphere in reinforcing droughtand desert-like conditions is correct.

Another example of the feedback mechanism is related to the irrigation ofagricultural land in arid and semi-arid regions. Poorly-designed irrigation schemeswith poor drainage may result in waterlogging and in soil salinization. This is verydetrimental to most plant species and may result in vegetation cover disappearance,increased albedo and wind blown dust, which in turn influences the surfacemicroclimate and causes changes in the rainfall regime and ecology. To be able tocontrol and reverse desertification, therefore, a better understanding is needed ofthe interdependency between resources-uses-changes and the postulated positivefeedback mechanisms.

The main point here is to emphasize that climatic vagaries, manifested in termsof drought, and human activities may combine to fuel the process of desertification.Desertification occurs first in drought years in patches of specific soil and vegetationtypes where a fragile equilibrium is initially, even if temporarily, disturbed. Thusdrought administers a shock to the ecological system.

2.2.7DESERTIFICATION AND FEEDBACK

MECHANISM

Over-exploitation of groundwater


Desertification is occurring everywhere in the world and even the most highlyindustrialized nations are not exempt from it. The United States, for example, hashad to face the degradation of its irrigated lands and increasing problems connected with its groundwater reserves. Russia faces similar problems on its irrigated lands. In Mediterranean Europe, forests and woodlands have been broughtinto better condition. In Australia, the advance of desertification has been halted.

It is in the developing world that desertification is accelerating most rapidly. Ithas been established that there is a negative correlation between desertification anddevelopment. The lack of success in arresting desertification in developing countriesis due to destructive land use practices and neglect of coherent planning.Unrestrained population growth and a corresponding rise in livestock numbersimpose increasing pressure on a fragile environment. Chronic shortages of food andthe desperate compulsion of these countries to export their crops against adverseterms of trade are continually destroying the fundamental resource base of thesecountries. The irony is that many of the urban problems of developing countrieshave their origins in the countryside. Loss of land productivity has forced villagersinto the towns.

Paradoxically, people living in areas strongly affected by desertification displaythe highest birth rates in the world. Desertification produces poverty. The decliningproductivity of dry zone croplands and rangelands has its most serious impacts onscores of millions of the world’s most wretchedly poor. Depending quite directly onthe soils and forests around them, people in the zones see their prospects for a betterlife dry up along with the natural resource base. Poverty in turn producesdesertification and the fight against mismanagement of the land can succeed onlyas part of a more general attack on underdevelopment. Desertification is not adetached technological problem; in most cases the term describes the ecologicaldimensions of a development process gone bad, a process which fails to providepeople with a reasonable standard of life.

Various developing countries have since accorded high priority in theirdevelopment plans to (i) self-sufficiency in food production; (ii) improved livingconditions for the people; and (iii) the restoration of ecological balance (Dhir,1986). Though countries like India have made major strides, particularly on the foodfront, the overall picture in developing countries still appears dismal. Despitenational concern and some action plans to combat desertification, the struggleagainst land degradation is getting harder and harder.

Because desertification is a phenomenon dynamic in time, space and intensity,its monitoring is an absolute necessity. The principal objectives of monitoringare to:

(a) Enlarge our knowledge and understanding of the processes, their causes and evolution;(b) Allow early detection of areas recently subject to accelerated desertification as a

result of climatic vagaries and the development of planning programmes.A number of biophysical and social indicators have been identified to quantify

desertification and monitor changes over time. For all of these indicators, we arerequired to know: (i) how widely occurring the particular process is spatially; (ii) the rate at which it is occurring. First of all, the rate of change in a particularcomponent is far from uniform, with many of them being highly episodic incharacter. For example, major advancement of sand sheeting and shifting sand dunescan occur during a few years of an extraordinarily strong wind regime. Likewise,major deterioration of natural vegetation may occur during periods of extendeddroughts with substantial recoveries in between. Therefore, in many of theseprocesses, short-term observations may lead to conclusions which are far off themean, if not misleading. Long-term studies are needed to discern a distinct trend inthe face of such large amplitudes in annual deviation. Secondly, the causative factor,i.e. human and livestock pressure, is never uniformly distributed spatially. A largevariation exists in the manifestation of the process even in situations where all otherland attributes remain the same. The establishment of the magnitude of a particularprocess therefore calls for observation at a number of sites (Gbeckor-Kove, 1988).

2.2.9MONITORING AND ASSESSMENT OF

DESERTIFICATION

2.2.8DESERTIFICATION AND

DEVELOPMENT


Analysis of rainfall and preparation of isohytal maps for a region, every 10 yearsfor example, will allow the displacement of isohytes towards the drier zone to befollowed, indicating progressive desiccation, and hence measures to avoid extendingrainfed operational agricultural activities into that region can be taken.

Global surveillance of the climate, the status of dryland ecosystems and of landuse can be achieved most economically through the remote sensing capability oforbiting weather satellites or earth resources satellites. In addition to sampling forground truth (e.g. vegetation changes) and aerial photography, satellite coveragecan also be used in conjunction with airborne remote sensing consisting of thermalinfrared cameras, multispectral visible and near infrared scanners and microwaveradars. Conventional air photography produces a finer resolution than satelliteimagery and is therefore useful for providing more detailed information about landuse in areas undergoing desertification. These methods can be used for mappingdesertification status.

The problems of desertification are multidimensional and require integration ofvarious areas of activity, such as settlement, development of natural resources ofsoil, water and vegetation, extension and training, research, etc. It is obvious thata lack of financial resources hampers integration and hence the tackling of theproblem effectively in developing countries.

As the degree of interaction of desertification processes varies from place toplace, counter-measures differ in importance from one affected country/region toanother. The reclamation of deteriorated lands requires cultural practices in order torecover the ecological equilibrium. These practices include the rational use ofnatural resources in arid and semi-arid areas, land use according to ecologicalrequirements and taking into consideration land potential and environmentalchanges. Methods of control aimed primarily at improvements in land use wouldalso improve the microclimate on which the natural vegetation and crops depend(Polevoy, 1992). Given the importance of economic development in developingcountries, afforestation and animal raising should be emphasized along with thecreation of a diversified economy. Obviously desertification can only be halted byencouraging multiple land use, exploiting the natural diversity of the biome and, atthe same time, exercising restraint in the size and scale of development projects in afragile ecosystem. Le Houerou (1987) discussed some myths and realities regardingthe prevention of desertification.

In view of the magnitude and complexity of the problem, it is important tohave a comprehensive time-phased plan of development with a clear idea of relativepriorities and the interdependence of various programmes. For instance, in dairydevelopment or sheep husbandry, it is necessary to produce sufficient fodder. Thiswill necessitate controlled grazing, the production of green fodder and the plantingof leguminous shrubs wherever possible.

The action programme to combat desertification is a major task which coversactivities in many disciplines and consequently encompasses several institutions,both in the spheres of investigation and implementation. It is also clear that thesolution to the complex problem of desertification must be executed simultaneouslywith other projects which have relevance to the desertification problem. Measuresto contain desertification (Polevoy, 1992; Zhu, 1990) should include integrating themeasures listed below:

(a) The natural resources of soil, water, forests and pastures should be judiciously utilized and conserved;

(b) Efforts should be made in the reclamation and distribution of new farms and to trainfarmers in modern agricultural methods to increase production and enhance income.

(c) There should be increased support for action-oriented research related to desertification and the improvement of affected lands;

(d) Greater regional and international cooperation in scientific, technical and socialsectors concerning desertification should be encouraged;

(e) Population growth should be controlled in order to reduce pressure on the land;(f) Energy efficiency in desertified areas should be improved to put an end to the undue

collection of fuel wood and the destruction of vegetation;

2.2.10RECOVERY AND CONTROL OF

DESERTIFICATION


(g) The livestock grazing pressure on pastures should be lowered and specialized foddergrowing farms created. In rangeland areas, proper rotational grazing should be practised;

(h) A preference for forestry to livestock breeding in the agricultural sector;(i) The protection of natural vegetation cover and the fencing off of deteriorated land

should be encouraged to allow better recovery conditions for grasses and other vegetation;

(j) The creation of protective forest belts to stabilize sand dune surfaces is an important element in the rational use of tree stands for sand arrest. The introduction of forest belts on lower sites and sand-binding species on sand dunesprovides an effective measure against desertification. The choice of suitable localspecies well adapted to the desert conditions is essential;

(k) The creation of a system of protective forest belts to fight wind and sand erosionnear oases may be practised. Such belts adequately protect field crops, arrest sandmovement and give food and shelter to wildlife; and

(l) Arid horticulture, including pomegranate trees, date palms and jujube, should beencouraged. Pasture improvement by introducing leguminous crops is yet anothereffective technique.

Desertification has four main causes; over-cultivation, overgrazing, deforestationand poor irrigation practices. These are human causes, the result of bad management of resources. Human and livestock population pressure often plays amajor role in desertification when the population exceeds the sustainable level infragile arid, semi-arid and sub-humid ecosystems. These are often exacerbated byother factors such as social and political systems which lead to unequal andinequitable access to resources, which force populations in developing countries tooverexploit the land for mere survival.

Natural causes of desertification, as well as those induced by people, are of anextremely complex nature and have varying forms, scales, duration and degrees ofintensity. Desertification is manifested through the degradation of plant and soilcover and depletion of water resources of the area.

Misconceptions are sometimes encountered concerning desertification and itsrelationship to drought. It is commonly felt that desertification spreads from a desertcore. Secondly, there is a belief that droughts are responsible for desertification,particularly in the semi-arid and sub-humid zones most frequently affected by severedroughts. The truth is that desertification can, and frequently does, occur far fromthe climatic Sahara, Kalahari and other deserts. The classic example of this isdesertification starting as a patch around a watering point from where it spreads outto nearby overgrazed rangelands. Furthermore, droughts do not cause desertificationbut they aggravate the harmful effects of improper land management so thatdesertification is intensified.

The problem of desertification is urgent because it results not only in the loss ofa nation’s productive resource base but also in the loss of valuable genetic resources,an increase in atmospheric dust with consequent changes in the radiation balance ofthe Earth and disruption of water resources.

The appropriate practical approach to combat desertification is one in whichthe peculiar characteristics of sensitive arid and semi-arid lands are taken into properaccount. This calls for the capability to measure changes and adequate support thatcan adapt easily to unexpected calamities such as a protracted drought (Sabadell,1982). In addition, development and management of resources should take intoconsideration the local human and natural resources, their diversity and their socio-economic needs. Such a comprehensive systematic approach would make themeasurement and assessment of cause-effect relationships, cumulative stresses anddetection of progressive land degradation possible.

2.2.11SUMMARY AND CONCLUSION


There are a number of practical suggestions and recommendations that can bemade to combat desertification around the world.

(a) Drought monitoring centres should be strengthened so as to have the capacity andcapability to issue warnings and long-term drought predictions. This calls for adequate data banks and computing power, trained personnel and a good telecommunication system;

(b) Efforts should be made to automate weather data collection by having a telemetered network of mini weather stations receiving on-line data from differentregions of the country;

(c) Action programmes to be set up to overcome drought situations based on theunderstanding of the causes and overall destructive effects of drought on the peopleand the area. Organizational capabilities and availability of resources should betaken into consideration while planning and initiating the programmes. Theinvolvement of local people and their organizations should form the key to theapproach;

(d) People must be taught that drought is a recurring phenomenon and hence theyshould prepare themselves to minimize its impact. This may include the cultivationof drought-resistant and durable varieties of crops;

(e) Appropriate contingency measures should be undertaken in the agricultural sectorto minimize crop losses. Measures should be taken to provide adequate fodder andnutrients for cattle;

(f) Water budgets should be prepared to optimize the utilization of water in reservoirsand groundwater resources;

(g) Steps should be taken to streamline the machinery for providing effective andtimely relief to drought-affected populations through the efficient implementationof relief measures;

(h) Action may be initiated for public health measures and for providing supplementary nutrition for young and needy children and expectant mothers indrought affected areas;

(i) Community assets should be created during the drought period which may includeafforestation and the planting of trees, social forestry, land reclamation, construction and repair of wells, tanks, reservoirs and ponds, the deepening ofexisting wells and water tanks to increase water reservoirs capacity, the creation ofminor irrigation facilities, small earthen and check dams for soil conservation, etc.;and

(j) Specific programmes should be initiated for combating desertification which mayinclude: (i) awareness of the dangers of indiscriminate felling of trees and over-grazing; (ii) forestry and agroforestry projects; (iii) pasture grass farming projects;(iv) poultry and fish farming projects; (v) dryland crop farming projects; (vi) small-scale irrigation and horticultural projects; and (vii) land conservation programmes.

• Baier, W. and Robertson, G.W., 1966: A new versatile soil moisture budget. Canadian Journal of Plant Science, 46:299–315.

• Barnett, T.P., Latif, M., Graham, N., Flugel, M., Pazen, S. and White, W., 1993:ENSO and ENSO-related predictability. Part I: Prediction of equatorial Pacificsea surface temperature with a hybrid coupled ocean atmosphere model. Journalof Climate, 6(1):545–566.

• Bruins, H.J., 1993: Drought risk and water management in Israel: Planning forthe future. In: Wilhite, D.A. (ed.), Drought assessment, management and plan-ning: theory and case studies, Kluwer Academic Publishers, Boston MA, Chapter8, 133–55.

• Bruins, H.J., 1994: Comparative chronology of climatic and human history inthe southern Levant from the late Chalcolithic to the Early Arab Period. In:Bar-Yosef, O. and Kra, R. (eds.), Late Quaternary chronology and paleoclimates ofthe eastern Mediterranean, Radiocarbon, Department of Geosciences, Universityof Arizona, Tucson, AZ and Peabody Museum, Harvard University, CambridgeMA, pp. 301–14.

2.2.12SUGGESTIONS AND RECOMMENDATIONS


REFERENCES

• Bruins, H.J. and Lithwick, H., 1998: Proactive planning and interactive management in arid frontier development. In: Bruins, H.J. and Lithwick, H.(eds.), The arid frontier: interactive management of environment and development,Dordrecht, Kluwer Academic Publishers, Chapter 1:3–29.

• Bruwer, J.J., 1993: Drought policy in the Republic of South Africa. In: Wilhite,D.A. (ed.), Drought assessment, management and planning: theory and case studies,Boston, MA, Kluwer Academic Press.

• Cane, M.A, Eschel, G. and Buckland, R.W., 1994: Forecasting Zimbabweanmaize yield using eastern equatorial Pacific sea surface temperature. Nature370:204–205.

• Charney, J. 1975: Dynamics of deserts and drought in the Sahel. QuarterlyJournal of the Royal Meteorological Society, 101:193–202.

• Chow, D., 1964: Handbook of applied hydrology. McGraw Hill.• Das, H.P., 1995: Incidence, impact, monitoring and mitigation of large scale droughts

in India. A review paper submitted to the Expert Meeting on DroughtMonitoring in WMO, Geneva.

• Das, H.P., 1999: Management and mitigation of adverse effects of drought phenomenon. In: Natural disasters – some issues and concerns. Natural DisastersManagement Cell, Visva Bharati, Shantiniketan, Calcutta, India.

• Dhir, R.P., 1986: Desertification processes in developing countries.WMO/FAO/IMD Workshop on Drought and Desertification in RA II and RA V, Pune, India.

• Doorenbos, J. and Pruitt, W.O., 1977: Crop water requirements. FAO irrigationand drainage paper 24. FAO Rome, pp. 144.

• Dregne, H.E., 1983: Desertification of arid lands. Harwood Academic Publishers,London, Paris, Utrecht, New York.

• Dyer, J.A. and Baier, W., 1979: An index for soil moisture drying patterns. Can.Ag. Engr. 21:117–118.

• Gbeckor-Kove, N., 1988: Drought and desertification. Lecture note on droughtand desertification in WMO training seminar in agrometeorology, TD No. 286.WMO Geneva.

• Gbeckor-Kove, N., 1995: Main climatic events affecting agricultural production.4ème Symposium International sur l’Assurance Agricole. Geneva, WMO.

• Gibbs, W.J., 1964: Space-time variation of rainfall in Australia. Symposium onwater use and management.

• Glantz, M.H., Katz, W., Magalhaes, A.R. and Ogallo, L., 1986: Drought followsthe plow (draft). National Center for Atmospheric Research, Boulder, Colorado.

• Glantz, M.H., 1987: Drought and economic development in Sub-SaharanAfrica. In: Wilhite, D.A., and Easterling, W.E. (eds.), Planning for drought:toward a reduction of societal vulnerability. Westview Press, Boulder, Colorado,U.S.A., pp. 597.

• Hare, F.K., 1983: Climate and desertification: a revised analysis. WCP44, WMO,pp. 149.

• Hare, F.K., 1984: Recent climatic experience in the arid and semi-arid lands.Desertification Control, 101:15–22.

• Hopkins, S.T. and Jones, D.E., 1983: Research guide to the lands of the world.ORYX Press.

• Hough, M., Palmer, S., Weir, A., Lee, M. and Barrie, I., 1996: The MeteorologicalOffice Rainfall and Evaporation Calculation System: MORECS, Version 2.0(1995), Hydrological Memorandum 45, updated. Available from the NationalMeteorological Library, Bracknell, Berkshire RG12 2SZ, UK.

• Ibrahim, F., 1978: Anthropogenic causes of desertification in western Sudan.Geo Journal, 2.3:243–254.

• Idso, S.B., Reginato, R.J., Hatfield, J.L., Walker, G.K., Jackson, R.D. andPrinter, P.J., 1980: A generalisation of stress day degree concept of yieldprediction to accommodate diversity of crops. Agricultural Meteorology,21:205–211.


• Issar, A.S., Zhang Peiyuan, Bruins, H.J., Wolf, M. and Ofer, Z., 1995: Impacts ofclimate variations on water management and related socio-economic systems. ParisUNESCO, International Hydrological Programme IHP-IV Project H-2.1.

• Jackson, R.D., 1982: Canopy temperature and crop water stress. In: Advances inirrigation, Hillel, D. (ed.), pp. 43–85.

• Katyal, J.C., Ramana Rao, B.V., Das, S.K., Victor, V.S., Vittal, K.V.R., SubbaReddy, G. and Vishnumurty, T., 1992: Forecasting of south west monsoon for1992: Management options for rainfed agriculture. Fert. News, 37(b):15–21.

• Kogan, F.N., 1997: Global drought watch from space. Bulletin of the AmericanMeteorological Society, 78:621–636.

• Le Houerou, H.N., 1987: Aspects meteorologiques de la croissance et du development vegetal dans les deserts we les zones menacees de desertification. TD No.194, WMO, Geneva, pp. 59.

• McCann, J., 1986: The social impact of drought in Ethiopia. In: Glantz, M.H.(ed.), Drought and hunger in Africa: denying famine a future, Chapter 11, Cambridge University Press, Cambridge, England.

• McKee, T.B., Doesken, N.J. and Kleist, J., 1993: The relation of drought frequency and duration to time scales. Proceedings, AMS Eighth Conference onApplied Climatology, 17–22 January 1993, Anaheim, CA, pp. 179–84.

• Mooley, D.A. and Parthasarthy, B., 1982: Fluctuations in the deficiency of thesummer monsoon over India and their effect on economy. Arch. Met. Geophy.Biol. Ser. B., 30 pp. 383–398.

• Mooley, D.A. and Parthasarathy, B., 1983: Variability of the Indian summermonsoon and tropical circulation features. Monthly Weather Review,111:967–978.

• Ogallo, L.J. and Gbeckor-Kove, N., 1989: Drought and desertification. TD No.286. WMO, Geneva.

• Oladipo, E.O., 1985: Desertification, its definition, causes, processes and relationship to drought. Paper presented at the WMO training seminar on the useof agrometeorological and hydrological data and information in the assessmentand combat of drought and desertification in Africa, Ethiopia.

• Olderman, L.R. and Van Lynden, G.W.J., 1997: Revisiting the GLASODmethodology, In: Lal, R., Blum, W.H., Valentin, C. and Stewart, B.A. (eds.),Methods for assessment of soil degradation, CRC Press, New York.

• Palmer, W.C., 1965: Meteorological drought. Research Paper No. 45. USADepartment of Commerce, Washington, U.S.A.

• Palmer, W.C., 1968: Keeping track of crop moisture conditions nationwide:The new Crop Moisture Index. Weatherwise, 21, No. 4.

• Parry, M.L. and Carter, T.R., 1987: Climate impact assessment: a review of some approaches. In: Wilhite, D.A., and Easterling, W.E. (eds.), Planning fordrought: toward a reduction of societal vulnerability. Westview Press, Boulder,Colorado, USA, Chapter 13.

• Pewe, T.L., 1981: Desert dust: an overview. In: Pewe, T.L. (ed.), Origin, characteristics and effect on man, Geological Society of America, pp. 1–10.

• Polevoy, A.N., 1992: Monitoring, assessment and combat of drought and desertification. CAgM Report No. 47, TD No. 505, WMO, Geneva.

• Ropelewski, C.F., 1995: Quantification of ENSO precipitation relationships.Journal of Climate, 7:1041–1059.

• Sabadell, J.F., 1982: Systematic approach to desertification. Paper presented toWorld Bank, Washington D.C.

• Sakamoto, C.M. and Steyaert, L.T., 1987: Toward a plan of action. Internationaldrought early warning programme of NOAA/NESDIS/AISC. Westview Press,Boulder, Colorado, USA, pp. 247–271.

• Sandford, S., 1979: Toward a definition of drought. In: Hinchey. M.T. (ed.),Botswana Drought Symposium, Botswana Society, Gaborone, Botswana.

• Sen, A., 1981: Poverty and famines: an essay on entitlement and deprivation.Oxford: Clarendon Press.


• Shafer, B.A. and Dezman, L.E., 1982: Development of a Surface Water SupplyIndex (SWSI) to assess the severity of drought conditions in snow pack run-offareas. In: Proceedings of the 50th Annual Western Snow Conference, pp. 164–175.

• Thiruvengadacheri, S., 1991: Satellite surveillance system for monitoring agricultural conditions in India. Unpublished paper. Drought management andpreparedness training seminar for the Asia and Pacific regions. Bangkok,Thailand.

• Torry, W.I., 1984: Social science research on famine: a critical evaluation.Human Ecology 12(3):227–252.

• Tucker, C.J. and Goward, S.N., 1987: Satellite remote sensing of drought con-ditions. In: Wilhite, D.A., and Easterling, W.E. (eds.), Planning for drought:toward a reduction of societal vulnerability. Westview Press, Boulder, Colorado,USA, Chapter 11.

• UNEP, 1984: General assessment of progress in the implementation of the plan ofaction to combat desertification: 1978–84, UNEP.

• Venkateswarlu, J., 1987: Technological and socio-political adaptation andadjustment to drought: an Indian experience. In: Wilhite, D.A., and Easterling,W.E. (eds.), Planning for drought: toward a reduction of societal vulnerability.Westview Press, Boulder, Colorado, U.S.A., pp. 391–408.

• Walker, J. and Rowntree, P.R., 1977: The effect of soil moisture on circulationand rainfall in a tropical model. Quarterly Journal of the Royal MeteorologicalSociety, 103:29–46.

• Warrick, R.A., 1975: Drought hazard in the United States: a research assessment,NSF/RA/E-75/004. National Technical Information Service, Springfield,Virginia.

• Watts, M., 1983: Silent violence: food, famine and peasantry in Northern Nigeria.University of California Press, Berkeley.

• White, D., Collins, D. and Howden, M., 1993: Drought in Australia:Prediction, monitoring, management and policy. In: Wilhite, D.A. (ed.),Drought assessment, management and planning: theory and case studies, KluwerAcademic Press, Boston, MA, Chapter 12.

• Wilhite, D.A. and Glantz, M.H., 1985: Understanding the droughtphenomenon: the role of definitions. Water International 10:111–120.

• Wilhite, D.A., 1986: Drought policy in the US and Australia: a comparativeanalysis, Water Resources Bulletin 22:425–438.

• Wilhite, D.A. and Easterling, W.E., 1987: Drought policy: Toward a plan ofaction. In: Wilhite, D.A., and Easterling, W.E. (eds.), Planning for drought:toward a reduction of societal vulnerability. Westview Press, Boulder, Colorado,USA, Chapter 37.

• Wilhite, D.A., 1990: Planning for drought: a process for state government. IDICTechnical Report Series 90–1, University of Nebraska, Lincoln, Nebraska,USA, pp. 52.

• Wilhite, D.A., 1992: Preparing for drought: a guidebook for developing countries.Climate Unit, United Nations Environment Programme, Nairobi, Kenya.

• Wilhite, D.A. (ed.), 1993: Drought assessment, management and planning: theoryand case studies. Kluwer Academic Press, Boston, MA.

• Wilhite, D.A. and Wood, D.A. (eds.), 1994: Drought management in a changingWest: new directions for water policy, IDIC Technical Report Series 94–1,International Drought Information Center, University of Nebraska, Lincoln,U.S.A.

• Wilhite, D.A. and Hayes, M.J., 1998: Drought planning in the United Statesand future directions. In: Bruins, H.J. and Lithwik, H. (eds.), The arid frontier:interactive management of environment and development, Kluwer AcademicPublishers, Dordrecht, Chapter 2:33–54.

• WMO, 1969: Estimation of maximum floods. TD No. 98, WMO, Geneva.• WMO, 1971: CAgM – V Session, October 1971, WMO, Geneva.• WMO, 1975: Drought and agriculture. TD No. 138, WMO, Geneva.• WMO, 1975a: Drought special environmental report No. 5. WMO No. 403,

Geneva, pp. 113.


• WMO, 1975b: Drought and agriculture. TD No. 138, WMO, Geneva.• WMO, 1983: Weather-based mathematical models for estimating development and

ripening of crops. TD No. 180, WMO, Geneva.• WMO, 1983a: Guide to hydrological practices, Volume II. WMO No. 168,

Geneva.• WMO, 1985: Application of agrometeorological and climatological data in the assess-

ment of and combat against desertification. Draft Special Environmental Report(unpublished), WMO, Geneva.

• WMO, 1992: Monitoring, assessment and combat of drought and desertification.Agricultural meteorology. CAgM Report No. 47, TD No. 505, WMO, Geneva.

• Yao, A.Y.M., 1981: Agricultural climatology. In: Landsberg, H.E. (ed.), Worldsurvey of climatology, general climatology, Volume. 3. Elsevier ScientificPublishing Co, New York, pp. 189–298.

• Zhu Zhenda., 1990: Desertification and its control in Northern China. In:Recent developments in geographical science in China, China Science Press,Beijing, pp. 57–67.


CHAPTER 3

INCIDENCE, PREDICTION, MONITORING ANDMITIGATION MEASURES OF TROPICAL CYCLONES ANDSTORM SURGES(by H.P. Das)

3.1 INTRODUCTION

Tropical cyclones are the off-spring of ocean-atmosphere interaction, powered byheat from the sea, driven by the easterly trades and temperate westerlies, the highplanetary winds and their own fierce energy. A tropical cyclone constitutes one ofthe most destructive natural disasters to affect many countries around the globe,causing tremendous loss of life, property, agriculture, etc. The impact of tropical cyclones is greatest over coastal areas which bear the brunt of the strongsurface winds and flooding from rainfall at the time of landfall. Besides the windswhich blow with lethal ferocity and heavy rainfall, cyclones are also associated witha devastating storm surge which inundates vast areas of coastline.

These intense low pressure systems develop over the oceans. Tropical cyclones,hurricanes and typhoons are regional names for what is essentially the samephenomenon. Depressions in the tropics which develop into storms are calledtropical cyclones in the south-west Indian Ocean, the Bay of Bengal and theArabian Sea, parts of the south Pacific and along the northern coasts of Australia;these storms are called typhoons in the north-west Pacific and are known ashurricanes in the Caribbean, south-east United States and Central America. In thePhilippines they are called bagious.

Compared with the extra-tropical cyclones, tropical cyclones are moderate insize. Still, their broad spiral base may dominate weather over thousands of squarekilometres, from the earth’s surface to the top of the tropical tropopause. Theassociated winds often exceed 200 km per hour, rainfall exceeds 50–100 cm in 24hours and, worst of all, very high storm tides (storm surge combined withastronomical tides) often exceed 6 metres, bringing the worst devastation over thecoastal areas where they strike. Today records are available for wind speeds of 317km per hour, gusting to 360 km per hour, rainfall of 117 cm in 24 hours and a stormsurge of almost 14 metres in association with tropical cyclones. The lowest pressureof 870 mb ever recorded was in association with a tropical cyclone (Typhoon ‘TIP’)in the Pacific which formed in 1979. Combined with duration, which is oftenseveral days, size and violence, the tropical cyclone is the most destructiveatmospheric disturbance (Smith, 1993). On a positive, though less dramatic side,tropical cyclones provide essential rainfall over much of the land they cross.

3.2 GEOGRAPHICAL DISTRIBUTION OF TROPICAL CYCLONES

A thorough knowledge of the climatology of tropical cyclones is essential forcyclone forecasting, risk assessment and for planning long-term mitigation measures for cyclone disaster management.

Almost all tropical cyclones form over the warm tropical waters (SST ≥ 27°C)of the tropical oceans except over the south Atlantic and south Pacific, east of140°W. Tropical cyclones are most commonly observed in the northern hemispherefrom May to November and in southern hemisphere from December to June. Thefrequency of tropical cyclones, however, varies from ocean to ocean. The annualaverage frequency ranges from 5.6 (the least) in the north Indian Ocean to about 30(the highest) in the north-east Pacific. About 70 per cent of tropical disturbanceswhich later develop into tropical storms form in the northern hemisphere.

Approximately 80 tropical cyclones attain maximum sustained winds of 20–25metres per second (m/s) over the globe per year. About half to two-thirds of thesedisturbances develop into the severe stage with a core of hurricane winds (windspeed 64 kts or more) (Gray, 1975). Figure 3.1 shows the annual frequency oftropical cyclones with surface wind speeds of 20–25 m/s over the globe.

The long-term average of tropical cyclones in the north Indian Ocean (the Bayof Bengal and the Arabian Sea) is 5.6 annually. This is about 6 per cent of the globaltotal. The frequency of tropical cyclones in the north Indian Ocean is the lowest inthe world. On average, 2–3 out of 6 tropical cyclones intensify to severe cyclones. Thefrequency of tropical cyclones is more in the Bay of Bengal than in the Arabian Sea.

(a) (b)

The average life period of tropical cyclones in the north Indian Ocean is about2.5 days as against the world average of six days which means, compared with otheroceanic areas, tropical cyclones in the north Indian Ocean are short lived.

The radial dimensions of tropical storms vary from a 50–100 km radius to a2 000 km radius. Over the Indian seas about 17 per cent of the storms havediameters between 3 and 5 degrees and 65 per cent between 5 and 10 degrees,indicating that the majority of storms have diameters within 1 000 km and aremoderate in size (WMO, 1993; Pisharoty, 1993).


Figure 3.1The global occurrence of

tropical cyclones over the globe(after W.M. Gray, 1975). Thenames of tropical cyclone bodiesand the tropical cyclones basins

covered by their programmes arealso indicated. The area

enclosed by the dottedlines is the area where the seasurface temperature is greater

than 26°C.

Figure 3.2Monthly distribution of

cyclonic disturbances andcyclonic storms for the

period 1891–1989(a) Bay of Bengal and

(b) Arabian sea

3.3 REGIONAL CATEGORIZATION OF TROPICALCYCLONES AND THEIR INTENSITY

Tropical cyclones are low pressure systems or depressions around which the air circulates in an anti-clockwise direction in the northern hemisphere and in aclockwise direction in the southern hemisphere. In the context of disaster prevention and preparedness, interest is mainly concentrated on those tropicaldepressions around which the wind blows with speeds exceeding 17 m/s (61 km perhour) near the surface. Meteorologists distinguish between tropical cyclones inwhich the wind strength is in the range 17–32 m/s (61–115 km per hour) and thosein which the wind speeds are greater than 32 m/s.

Tropical cyclones, as pointed out earlier, are known by different terms accordingto the regions in which they occur. These descriptive terms in relation to wind speedare given in Table 3.1 (ESCAP/WMO/LRCS, 1977).

It is important to be aware of the different regional names given so that, forexample, it will be appreciated that what is described as a severe cyclone in the Bayof Bengal is essentially the same phenomenon as that which is called a hurricanewhen it occurs in the North Atlantic.

3.4 CHARACTERISTICS OF A TROPICAL CYCLONE

The tropical cyclone is a severe type of weather system, having distinct characteristics (Gray, 1968; Frank, 1982):

(a) They are neither associated with moving anticyclones nor with fronts;(b) They form only in certain regions of the tropics;(c) No regularity exists in their formation or movement;(d) They form in the ocean where the surface temperature is above 27°C;(e) Pressure distribution and other properties are fairly symmetrically distributed

around the centre with a nearly circular isobar;(f) Though they derive their energy from the latent heat of condensation for their

generation as well as for their sustenance, the precipitation amount varies fromstorm to storm;

(g) They occasionally have a central sea level pressure of 900 hPa or lower and surfacewinds exceeding 100 kts;

(h) They exist only over oceans and weaken rapidly on crossing the land; and(i) The intense system often has a core of calm or very light winds. This region is

known as the “eye” of the storm which has on an average a diameter of 20 kms andis free of any weather.

3.5 CONDITIONS NECESSARY FOR TROPICAL CYCLONE FORMATION

Based on observational evidence of a large number of storms in the Pacific andAtlantic oceans, the following environmental factors favourable to cyclones wereidentified (Gray, 1968; Anthes, 1982):

(a) Large values of low level relative vorticity;

CHAPTER 3 — INCIDENCE, PREDICTION, MONITORING AND MITIGATION MEASURES OF TROPICAL CYCLONES AND STORM SURGES 37

Range of maximum wind speedsRegion (metres per second)

17–32 32–85

Western North Pacific Ocean Tropical cyclone TyphoonBay of Bengal and Arabian Sea Cyclone Severe cycloneSouth Indian Ocean Tropical depression Tropical cycloneSouth Pacific Ocean Tropical depression CycloneNorth Atlantic Ocean and Tropical storm Hurricaneeastern North Pacific Ocean

Table 3.1Areas of occurrence and

regional description

(b) Weak vertical wind shear of horizontal winds;(c) Large value of Coriolis parameter;(d) Sea surface temperature exceeding 27°C and a depth of warm water;(e) Degree of convective instability; and(f) Large values of relative humidity in the lower and middle troposphere.

3.6 STORM SURGE

A storm surge is the abnormal rise of the sea level caused by the movement of thecyclonic storm over a continental shelf. It is caused by the pressure drop near thestorm centre and the surface drag due to the strong winds accompanying the storm.Storm surge is the most devastating feature associated with a tropical cyclone. Mostloss of human life and cattle is due to storm surges (Das, et al., 1974; Pisharoty, 1993).

The storm surge depends on the following factors:(a) Intensity of the system;(b) Landfall point and time;(c) Offshore bathymetry near landfall and its latitude;(d) Radius of the maximum wind;(e) Speed of the storm; and(f) Angle of the track relative to the coastline.

The effect is most pronounced where the sea is shallow. The actual tide is thesum of the astronomical tide and the storm surge and as such it is highest at the timeof high tide. This occurs only at the coast near the landfall point – highest to theright and lowest to the left of the point of crossing. Usually the rise is 2–3 metres,but can be as much as 5 metres in the case of a severe cyclone. A surge may alsoinvade inland along a major river as a tidal wave and can cause flooding even 10 or15 kilometres upstream. This causes saline inundation of croplands 10–15 km insidethe landfall, inflicting irreparable loss to the coastal economy.

The storm surge occurs some distance away from the centre of the tropicalcyclone, to the right of centre in the northern hemisphere and to the left in thesouthern hemisphere. One of the main forecasting problems, therefore, is to estimatewhere and when the centre of the storm will make its landfall. This estimate is apreliminary step to predicting the probable height of the storm surge at variousplaces along the coast. Ghosh (1977) developed a dynamic method to estimatestorm surge along the entire east coast of India where most of the severe cyclonesfrom the Bay of Bengal cross the coast.

Coastal embankments susceptible to storm surges should be designed specifically towithstand the expected storm surge water heights and forces, the combined actionof wind and waves and overtopping from the storm surge water. Furthermore,coastal embankment projects in deltaic areas should be planned in conjunctionwith other development projects, such as highways and harbour and reclamationprojects, in order to avoid duplication of investment costs (Basu, 1999).

An advanced and carefully planned system of storm surge protection has beendeveloped in Osaka, Japan, in one of the most densely populated areas of Asia. Thepresent storm surge prevention project consists of embankments, locks, pumpingstations and raised bridges lowered by land subsistence. The large locks that nowprotect Osaka from a storm surge caused by typhoons were constructed as analternative to raising the height of the existing embankment for reasons of cost andtime and to minimize the disruptive effect on traffic.

3.7 HEAVY RAINS ASSOCIATED WITH A HURRICANE/TYPHOON

The first manifestation of an approaching tropical cyclone is torrential rain. Thismay occur far from the centre of a hurricane and often starts around 12 hours beforelandfall. Three rainfall bands are normally formed in a hurricane:

(a) the eyewall region;

3.6.1PROTECTION FROM THE

STORM SURGE


(b) the spiral rain bands; and(c) the outer heavy rain region.

The eyewall region is normally within an annular region of strong ascendingmotion and strong precipitation about 15–50 km from the hurricane centre. Thespiral rain band is usually located in the right semi-circle of a tropical cyclone inthe northern hemisphere (the left semi-circle in the southern hemisphere). Thesecan produce torrential rain. The outer heavy rain region is normally located to thenorth of the cyclone centre (in the northern hemisphere) where an inverted troughis formed (Raghavan, 1997; Krishna Rao, 1997).

Heavy rainfall is normally influenced by the following factors:(a) Sustenance of the hurricane vortex after landfall;(b) Stagnation of the hurricane;(c) Sustained supply of water vapour;(d) Interaction between middle and lower latitude circulation;(e) Topographical effects; and(f) Small and mesoscale circulation systems in the hurricane.

Rainfall over land commences when the centre of the cyclone is about 500 kmaway from the coast. The rainfall continues along and near the track of the cycloneeven after it crosses the coast and moves over the land as a weak system and finallydissipates (Pisharoty, 1993).

During this period an average rainfall of 5 cm occurs over a land area about2 000 km long and about 500 km broad. The amount of water thus precipitated is50 000 million cubic metres (1 cubic metre = 1 000 litres).

In several cases, rainfall of 40 cm in one day can occur over an area of about50 km radius around the centre of the cyclone. It can be 20 cm a day even atdistances of 200 km from the centre. Usually the rainfall distribution around thecentre is not symmetric; more rainfall occurs to the left of the track than to theright. Less, but still heavy, rainfall occurs even at distances 400 km away from thecentre. Beyond that some rainfall occurs, although it is of the order of 1 or 2 cm aday. The weakened cyclone moves across the country for two or three days, givingcopious rainfall.

The intensity of rainfall is quite high particularly within the core of hurricanewinds. It can reach 10–12 cm per hour. Outside the core heavy intensities of 4–6cm also occur, but they are over smaller areas, 1 00 square kilometres or so in size,and for a shorter duration of an hour or so. As the centre of the cyclone strikes thecoastline, the strong winds and the heavy rain inflict havoc on the land through soilerosion, even over those areas not subjected to the storm surge. A cyclonic stormafter crossing the coast may get a fresh supply of moisture and continue to causeheavy rainfall. In India similar situations often occur due to a fresh supply ofmoisture from the Arabian Sea.

3.8 SURFACE WIND IN A TROPICAL CYCLONE

Strong winds associated with tropical cyclones are mainly responsible for the hugeloss of human life, cattle and property. Most deaths occur due to houses collapsingand the uprooting of trees. The strength of the wind depends upon cyclone inten-sity, structure of the vortex and the environmental pressure distribution.

The band of maximum winds associated with a typical tropical cyclone is20–50 km away from the storm centre. However, this band may vary from between10–150 km away. The distribution of maximum winds is also highly asymmetricalwith respect to the storm centre. This may result from environmental forcing, innercore restructuring or the interaction between the moving tropical cyclone and theunderlying surface. In India, the following modified Fletcher’s formula, suggested byMishra and Gupta (1976), is used for estimating maximum wind speed:

Vmax = 14.2 Pn – P0

where Pn – P0 is the pressure depth.


Dvorak’s (1975) relationship between a satellite-derived T-number and MSLP isalso very useful in estimating the Vmax. This has been widely used in cases of non-availability of ships’ observations. With an increase in the severity of a tropicalcyclone, its damage potential also increases.

3.9 AVAILABILITY OF DATA FOR MONITORING AND FORECASTINGTROPICAL CYCLONES

The monitoring and forecasting of tropical cyclones are dependent upon data covering a wide area, on telecommunication facilities to allow the data to be collected from the numerous observing stations and on the data being broadcast forinterpretation by all the services requiring it (ESCAP/WMO/LRCS, 1977).

In order to provide a forecast for an area the size of a country or continent, theforecaster needs data giving conditions at the surface and in the upper atmosphereover a large area. Because atmospheric processes are taking place all the time, thedata must be constantly renewed, some reports being required at hourly intervals,others every 3, 6 or 12 hours, so that the state of the atmosphere and of the seasurface can be monitored continually. Some observing facilities, e.g. weather radar,permit continuous surveillance of rain and cloud within range of the equipment.Information made instantly available in this way is highly valuable.

A country susceptible to tropical cyclones should install additional observationfacilities to supplement the basic meteorological network required for its normalforecasting and climatological requirements for aviation, industry, agriculture,shipping, the general public, etc. The additional facilities required by a countryvulnerable to tropical cyclones should be developed on the following lines:

(a) Weather radarsThe unique and virtually indispensable advantages of a weather radar are that,within its range, it provides a continuous watch on a tropical cyclone and enablesthe meteorological service to provide reliable, accurate information as the stormcomes closer and closer to the coast.

(b) Auxiliary reporting stationsThese stations should be equipped to measure pressure, wind and rainfall. Theyshould be deployed along coastal areas and at important locations inland.

(c) In-flight reports from aircraftReports from commercial or other aircraft are always of great value in providingdata from areas remote from the standard observing network, e.g. over the ocean.When it is known that a tropical cyclone has formed, requests for the transmissionof in-flight weather reports should be made to any air crew expected to fly in thevicinity of the storm.

(d) Aircraft reconnaissance reportsUS reconnaissance aircraft penetrate hurricanes in the North Atlantic andtyphoons in the Pacific. These flights provide valuable meteorological informationincluding the position of the centre, reports on cloud structure and on the distribution of temperature, wind and pressure.

A national meteorological service requires an elaborate telecommunications systemto collect and retransmit data obtained from the national network. As and whenthe cyclone is formed and moves, the national meteorological telecommunicationssystem, which includes land line and radio between countries in the vicinity of thecyclone, should, in addition to exchanges of plain language and numerical codeddata, provide for the reception and transmission of weather charts by facsimile andfor receiving pictures from the meteorological satellites.

Telecommunications facilities

Observational data


3.10 DESTRUCTION CAUSED BY TROPICAL STORMS

The tropical cyclone causes irreparable damage to the agriculture, ranches andforests. The loss to an agriculture system can be categorized as follows:

(a) Destruction of vegetation, crops, orchards and livestock;(b) Damage to irrigation facilities such as canals, wells and tanks; and(c) Long-term loss of soil fertility from saline deposits over land flooded by sea water.

The loss caused by a single storm may run into millions of dollars (Holland andElsberry, 1993). This is particularly so in the case of developing countries. Coastalareas in developing countries suffer great loss of life, especially the Indo-Bangladeshcoast where the shallow bathymetry of the north Bay of Bengal is prone to thehighest storm surges in the world. These areas are also densely populated and arecentres of brisk marine activity. Most of the population dwell in temporary thatchedhouses and the farmers have small land holdings. The lack of cyclone shelters andproper escape routes, the slow mode of transportation and the low elevation of theestuarine area all contribute to the regular catastrophes which occur here.

The effects of strong winds in coastal areas are seen in stunted and often verysculpted trees providing unmistakable evidence of the direction of the strong winds.In addition to the battering effects of winds, there is the additional damage causedby airborne sea salt which occurs within a few hundred metres of the coast. Winds which blow from coastal seas spray a lot of salt on coastal areas, making itimpossible to grow crops sensitive to excessive salt. Tamate (1956) details the resultsof work in Japan on reducing the salt content of the air (by filtering) by shelter belts.Immediately to the lee of shelter belts, salt concentration can be lowered to 12 percent of that to the windward side.

Moreover, a rise in the ecstatic sea level results in the territorial extension ofcoastal salinity under the direct or indirect influences of sea water. Fields inundatedby the storm surge suffer a loss of fertility due to salt deposition, even after the seawater has receded. The affected land takes a few years to regain its original fertility.The period of high water can last from 6 hours to several days, if the drainage ispoor, and may leave the soil saline and unfit for crops. Saline soils are predominantlyobserved in the coastal areas of India. A rise in the sea level would adversely affectthe 7 000-km coastal belt of India, comprising 20 million ha of coastal ecosystem,increasing coastal salinity and reducing crop productivity.

For most developing countries, agricultural production losses represent a significantpart of the damage caused by cyclonic disasters. The amount of damage caused toagriculture (and forests) by the high winds depends on the velocity of the winds andtheir duration. The higher the wind speed and the longer the duration of the strongwinds, the greater the damage. Typhoons have been known to inflict severe damageon agriculture: for example, in south Hainan on 2 October 1989, some 25 milliontimber and rubber trees were blown down (Salinger, WMO, 1994) and in Mauritiuson 6 February 1995 the main agricultural product, sugar cane, was reduced by 30 percent. A typhoon which struck Thailand on 4 November 1989 wiped out some 150 000ha of rubber, coconut and oil palm plantations and other crops (WMO, 1997).

In Mozambique, it was reported that more than 100 people died while 30 000others were affected by the cyclone which struck the country in January 1984. Thetotal cost of damage to agricultural crops was estimated at US$ 75 million. Thecrops which suffer most in Thailand are rice, upland crops and fruit; total area lossescomprise nearly 160 000 ha of these commodities annually. At the same time, thelogging industry suffers financial losses of US$ 30–450 million and areal losses of4 800–7 6000 ha annually. In the USA, annual crop losses are approximately US$50 million. In the Philippines, livestock losses of about $4 million (for 1991) andcrop losses of about US$ 5 million (for 1992) were reported. In Vanuatu, cropsaffected annually are coconuts (US$ 165 000–826 000 losses) with a damaged areaof 50 000–100 000 ha; cocoa (US$ 41 000–226 000) with a damaged area of10 000–20 000 ha and garden crops (US$ 816 000–4 110 000) with a damaged areaof 500–1 000 ha (Bedson, 1997).

3.10.3AGROMETEOROLOGICAL LOSS

ASSOCIATED WITH SOME

DEVASTATING CYCLONES

3.10.2SALT DEPOSITION IN

COASTAL AREAS

3.10.1DAMAGE TO AGRICULTURE


Traditional, small-scale fisheries are also hit by cyclones. In monetary terms, thelosses incurred by livestock raising, forestry and fisheries mostly remain below thosesuffered by crops. In Madagascar, following several cyclone occurrences in 1983–84,the FAO Office for Special Relief Operations (OSRO) estimated that crop lossesrepresented 85 per cent of the total damage to the agricultural sector, whereas thedamage to infrastructure and equipment (drainage and irrigation channels, fishinggear, etc.) barely reached 15 per cent. Livestock losses were negligible (OSRO,1984). A rather different impact pattern occurs in small islands like Antigua andBarbuda where fisheries constitute the backbone of the economy. After hurricaneHugo in 1989, 47 per cent of the losses occurred in fisheries, but crop losses stillrepresented almost 40 per cent of the total damage (OSRO, 1989).

It is worth mentioning the losses affecting cash crops which are a major sourceof export earnings for a number of developing countries. In Nicaragua, it is reportedthat direct loss of export crops due to hurricane Juana (Joan) in late 1988 amountedto 21 per cent of total losses in the agricultural sector (MIDINRA, 1988). Coffeeand bananas suffered a direct loss of their fruits and mechanical damage to plants.Nonetheless, food crop losses were estimated to be higher (35 per cent), while thelivestock sector was less affected (8 per cent), of which one fifth was poultry.

Two broad categories of effects on the agricultural sector can be identified:direct and indirect effects. Direct effects to a farmer could be, for example, the lossof his current crop and damage to his irrigation facilities. Indirect effects appearprogressively, as a result of low income, decrease in production and other factorsrelated to the cyclone disaster. The farmer may well have to pay high prices for seedsbecause of increased demand and disruption of the transportation system. He mightalso lose a portion of his future harvest because of storm surge-related salinizationof soil or the destruction of perennial plantation crops which sometimes take 5–10years to re-establish. Indirect effects are thus difficult to quantify and therefore areoften termed “invisible” effects. Conditions conducive to the development of pestsand diseases are to be regarded as indirect effects. Tropical examples are desert locustoutbreaks or increased disease incidence in sugar cane after hurricanes. Plantsweakened by adverse weather are much more susceptible to cryptogamic diseases orpest attacks, such as the explosion of coconut black beetle on wind-damagedcoconut trees. For example, it is taking 30 years for certain timber varieties to re-establish after hurricane Allen hit Jamaica in August 1980 (FAO, 1982). It isfrequently from 6–12 months for banana and 4–5 years for coffee and sugar cane.Even for crops that regenerate easily after partial damage, harvesting is usually madedifficult by the “abnormal” morphology, thus further reducing yield.

In November 1970, one of the worst cyclones in history struck Bangladesh. Thiscoincided with the start of the Aman rice harvest, of which a sizeable proportioncould not be harvested. This led to a drop in production of 1 million tons ascompared with the previous year. Moreover, November is also the planting periodfor the Boro rice crop. Following the cyclone and ensuing floods (including a tidalwave), there was great disruption to economic activity, destruction of infrastructureand the area cultivated with Boro rice was reduced and planting or transplantingwas delayed. This accounted for the sharp decrease in Boro production the followingyear (Gommes, 1997).

Between 14 and 16 November 1991, tropical storm 4B pounded southern Indiafor two days; sea water inundated the coastline resulting in extensive property andcrop damage. Several irrigation tanks were breached and crops on 2.36 lakh ha weredamaged in Tamil Nadu State and 7 253 head of cattle and 4 500 poultry birdsperished in Andhra Pradesh State, apart from damage to crops worth crores ofrupees. In mid-November 1992 cyclone 10B brought in its wake extensive damageto life, property and agricultural production in the coastal districts of Tamil Naduand Kerala in India and in Sri Lanka. Hundreds of thousands of hectares ofagricultural land were submerged and thousands of cattle perished in Tamil Naduand Kerala states. In March 1992 cyclone Fran hit the Queensland coast causingflooding and damaging sugar cane crops. In central Vietnam, tropical storm Angela(1992), accompanied by gusty winds and heavy rains, caused considerable damage torice and vegetable crops (WMO/UNEP, 1993).


On 7–8 August 1993, tropical storm Bret (FAO, 1994) affected Venezuelacausing agricultural losses, according to official sources, of 14 000 ha of maize (22per cent of the area under cultivation), 1 500 ha of rice, 750 ha of cassava, 150 ha ofbananas, 150 ha of horticulture/fruit trees, as well as 2 000 head of cattle, and largenumbers of pigs, poultry and other animals.

Another example of a natural disaster was that of cyclone Nadya which affected13 of the 22 districts in Nampula, Mozambique, in early April 1994, about twoweeks before the harvest. Food crops such as maize, cashew, maxdeira, mapira andcassava were badly affected. In the worst affected districts, 80 per cent of these cropswere lost and about 870 000 people were considered to be affected. In addition to itsaffect on agriculture, the cyclone caused outbreaks of diarrhoea and choleraassociated with unsanitary conditions (DHA, 1994).

During May 1986 tropical cyclone Namu caused 100 deaths and nearly one-third of the population to leave their homes in the Solomon Islands in the southPacific. Agricultural losses, especially timber, cocoa, coconuts and palm oil, wereparticularly heavy. It was estimated that 10–15 per cent of oil palm, 15–20 per centof copra and 10–25 per cent of cocoa production would be lost over the subsequentthree years (Britten, 1987).

Tropical cyclones are also responsible for a large number of casualties and considerable damage to property. Destruction is confined to the coastal districts,the maximum destruction being within 100–150 km of the centre of the cycloneand to the right of the storm track where wind direction is from ocean to land. Theprincipal causes of destruction by cyclones are: (i) fierce winds; (ii) torrential rainand associated flooding; and (iii) high storm tides (the combined effect of stormsurge and astronomical tides) leading to coastal saline inundation.

Most casualties are caused by coastal inundation by storm tides; penetrationvaries from 10–20 km inland from the coast. Heavy rainfall and floods come nextin order of devastation. Death and destruction purely due to winds are relativelysmall. The collapse of buildings, the falling of trees, flying debris, electrocution anddisease from contaminated food and water in the post-cyclone period contributesubstantially to loss of life and destruction of property. Available statistics the worldover show that the tropical cyclone is far ahead of any other disaster as a killer,accounting for about 64 per cent of total lives lost. The 80–100 tropical cyclonesthat occur each year caused an annual average of 20 000 deaths between 1964 and1978. The average economic loss per cycle per year was about US$ 60 million.When a cyclone hits the US coast, for example around the Gulf of Mexico, the losscan be as high as US$ 2 000 million (ESCAP/WMO, 1991).

Apart from the serious calamity of loss of human life and injuries, the impacts ofa severe cyclone on a coastal district are:

(a) Damage to fishing and other facilities;(b) Damage to off-shore and on-shore installations;(c) Damage to roads, railway tracks and other public utilities;(d) Damage to electricity supply systems; and(e) Damage to telecommunication systems.

The factors discussed above – damage from wind, rain, flood, storm surge and seawaves – may be regarded as representing the direct impact of tropical cyclones. Lossand damage attributable to these factors can be assessed in terms of deaths andinjuries to the population, buildings and installations destroyed or damaged,destruction of crops and livestock, etc. However, there are additional, perhaps indirect, consequences which cause losses to individuals, industry, the communityand the nation. The magnitude of these effects can be very large and they cannotbe ignored. Some brief discussion of these aspects follows.

A tropical cyclone can lead to disruption of the workforce and to other activitiesresulting in a substantial loss in productivity. Factories and warehouses may be out ofcommission for a time and many working hours may be lost because of breakdownsin land, sea and air traffic, impeding the movement of people and supplies, and

Losses in productivity

3.10.5SOME ECONOMIC AND SOCIAL

CONSEQUENCES

3.10.4OTHER DESTRUCTIVE EFFECTS OF

CYCLONES


because of a diversion of labour to assist with disaster relief and restoration. Inagriculture, there can be large losses in primary production on account of delays inthe recovery of arable land that has been inundated.

A tropical cyclone can cause many losses of a personal and domestic nature. Theloss of personal belongings, such as clothing and furniture, can be an especiallysevere blow to families whose financial reserves are small. In the domestic area,breakdowns in public utilities can lead to significant losses. All these losses may begreat in some homes and small in others but represent a substantial financial loss tothe community as a whole.

3.11 BENEFICIAL IMPACT OF CYCLONIC STORMS

The effects of cyclonic storms are not wholly bad; benefits centre principally on theprecipitation associated with them which may have considerable value to agriculture. The heavy rain associated with cyclonic storms guarantees a longerperiod of water availability and provides possibilities for off-site extra storage inrivers, lakes and artificial reservoirs (on farms or at the sub-catchment level) givingan improved rural water supply and expanded or more intensive irrigated agriculture and inland fisheries. The extra precipitation due to cyclonicstorms on land helps to increase plant growth improving the protection of the landsurface and increasing rainfed agricultural production. Ryan (1993) mentioned some important benefits of tropical cyclones in Australia. Increasedwater availability in water-critical regions makes agricultural production less susceptible to the dry period. Sugg (1968) estimates that nine major hurricanes inthe United States since 1932 terminated dry conditions over an area about 622 000square kilometres (240 000 square miles). Hartman, et al. (1969) estimated thechange in total crop value brought about by these storms occurring in differentmonths. The losses in crops for two of the storms were $54 million and $1 million;for the third storm there was an increase in total crop value of $8 million.

3.12 CYCLONE WARNING SYSTEM

One of the short-term mitigation measures against tropical cyclone disaster is anefficient cyclone warning system. The requirements for an efficient cyclone warning system are:

(a) Advance, accurate and detailed forecasts of dangerous conditions;(b) A rapid and dependable distribution system for the forecasts, advisories and

warnings to all interested parties; and(c) Prompt and effective utilization of warnings by the government and the public.

To have a fairly accurate forecast, it is necessary to have: (i) maximum high-quality data; (ii) forecasters with sufficient ability, training, experience and time fordata preparation; and (iii) foolproof techniques for preparing accurate forecasts ofthe storm’s movement, changes in intensity and storm tides. The US WeatherResearch Program for Hurricane Landfall (OFCM, 1996; Elsberry and Marks, 1999)promises improved forecasts of track, intensity, surface wind and rainfall, as well asresearch on decision-making and the technology transfer necessary to convertadvances in science and technology into products useful to society. Thecommunication system for the distribution of advisories and warnings should be onethat can dependably deliver the advisory information to all concerned in theshortest possible time, even in cyclonic conditions of strong wind, heavy rain, floods,etc.

An essential element of a warning service is that there should be certainty thatthe warnings will reach the intended recipients promptly. The supportingcommunications system, including back-up facilities, should therefore be plannedand implemented in full detail.

Personal and domestic losses


It is also essential that the responsible authorities and individuals who receivetropical cyclone warnings should be clear as to the actions that follow as soon as thewarning messages have been received at their end. The warning itself might be thesignal for prearranged action to be taken. This would, of course, be the first of aseries of executive measures that responsible authorities would set in motion(WMO, 1983). The meteorological warning, besides giving precise informationabout the tropical cyclone itself and the winds and rainfall to be expected, mightalso serve as a preliminary warning of a flood or storm surge. Such preliminarywarnings should be confirmed or amended in due course by the forecast centre, inconjunction with hydrographers in the case of a storm surge warning. There hasbeen considerable analysis of the key components of warning systems, for example,in the context of the WMO Tropical Cyclone Programme and the InternationalDecade for Natural Disaster Reduction (WMO, 1990).

For warnings indicating likely places of landfall, the associated maximum wind,rainfall and storm surge heights are essential for making preparations to save lifeand property, and it is absolutely essential to maintain an efficient telecommunicationssystem for the dissemination of warnings.

Cyclone warnings to the main users in India are issued in two stages. In the firststage a “cyclone alert” is issued normally 48 hours before the commencement ofadverse weather along the coast. In the second stage a “cyclone warning” is issuedaround 24 hours before the cyclone strikes the coast. Port and fisheries warningsstart much earlier. These warnings are disseminated through (i) landline telegrams ofspecial high priority; (ii) repeated broadcasts through All India Radio in differentlanguages; (iii) bulletins to the press; (iv) the Posts and Telegraph Department’scoastal radio stations (broadcast in code for the benefit of ships on the high seas);(v) telephone, telex and teleprinters wherever available; and (vi) the wirelessnetwork of the police.

3.13 DISASTER MANAGEMENT AND MITIGATION MEASURES

Accurate warnings of a tropical cyclone will have no impact unless protective measures are taken by the government and the affected people. Recently government awareness has increased greatly and proper action is being takenregarding tropical cyclone warnings. Disaster management refers to all activitiesconnected with prevention, preparedness and relief (Mandal, 1993). Disaster prevention may be defined as measures designed to prevent natural phenomenafrom causing or resulting in disasters. A disastrous event does not pose much of thethreat and ceases to be a disaster if suitable and adequate mitigation measures areadopted well in advance. Prevention of the formation of tropical cyclones is not inthe realm of possibility, but much of their disastrous potential can be reduced,restricting thereby the loss of human life and loss of property, by adopting appropriate strategies and taking timely precautions on the receipt of weatherwarnings. Climatological data helps in the advance preparation of long-range policies and programmes for disaster prevention. Disaster preparedness, on theother hand, is the action needed to minimize the loss of life and damage to propertyby organizing timely and effective rescue, relief and rehabilitation operations whenan area is struck by a disaster. Thus, disaster prevention is essentially based on climatology, and preparedness on weather warnings. Though different in nature,both require advance and complementary planning. Some important aspects of dis-aster management have been mentioned in ADB (1992).

Four ingredients are required to implement disaster prevention planning: (i) a technical evaluation of the climatological risk of cyclones and cyclonic effects inthe coastal areas; (ii) an assessment of the relative vulnerability of population,within the selected boundaries; (iii) the establishment of structural design codes,regulatory controls and minimum safety standards designed to encourage publicadherence; and (iv) an educational programme to gain community acceptance of

3.13.1PREVENTION

3.12.1DISSEMINATION OF CYCLONE

WARNING


the cost of cyclone disaster prevention. It is evident that long-range tropicalcyclone mitigation planning or prevention measures are highly interdisciplinarybecause they involve liaison between meteorologists, hydrologists, environmentalplanners, engineers, agricultural specialists, marine scientists and administrators.The objectives of tropical cyclone disaster prevention in national planning havebeen comprehensively described in guidelines in ESCAP/WMO/LRCS, 1977.

Disaster preparedness for impending cyclones, as we know, refers to the plan ofaction needed to minimize loss to human lives, damage to property and agriculture.The effectiveness of disaster preparedness ultimately depends on the effectivenessof planning and response at the district or local government level (Oakley, 1993).

“Preparedness” in an agriculture system can include early harvesting of crops,if matured, safe storage of the harvest, etc. Irrigation canals and embankments ofrivers in the risk zone should be repaired to avoid breaching. Beyond this, as thestorm approaches the area, nothing more can be done.

Examples of major non-agricultural-related decisions under “preparedness”include the evacuation of ships from ports, the progressive closure of sea, road, rail,air, and inland river transportation systems and power supplies, the closure schoolsand suspension of commercial activities, requests to the military for assistance withevacuation, emergency food, clothing and medical supplies, etc. Each of theseactivities requires a special meteorological briefing to explain the contents of officialor broadcast warnings. The cyclone warning centres need to liaise with the relevantradio stations for broadcasting the latest cyclone bulletins. The public and officialsare advised to monitor these weather bulletins. Fishermen in the concerned coastalbelt must be specifically warned not to go to sea. Irrigation and waterways authoritiesshould inspect the embankments of rivers in the risk zones to avoid breaching. Thisstage helps to avoid the danger of over-warning.

Twenty four hours before the likely onset of adverse weather, a specific warningabout the position of the storm, its intensity and likely place of landfall andassociated weather, including tidal waves, should be issued and given wide media(visual and audio) coverage.

The following are the important steps to be taken for “disaster preparednessdue” for cyclones:

(a) Adequate storm warnings for mariners on the high seas;(b) Port warnings for the safety of ships leaving port (distant signals);(c) Port warnings for the safety of the port, ships and craft plying coastal areas and ships

moored in the port;(d) A warning for the safety of fishermen; and(e) A cyclone warning to the state government authorities and non-government

agencies for safety of the coastal population.

In many coastal areas it may be prudent to evacuate the population in advance ofa storm. Evacuation can be horizontal or vertical. The former makes use of routesto move people to higher ground inland, the latter moves people to highrise, storm-proofed buildings. Both types of evacuation require careful contingency planning,identifying routes and possible hazards (ESCAP/WMO/LRCS, 1977). There aremany problems specific to horizontal evacuation. For example, routes inland maybecome blocked by pre-storm flooding, fallen trees and power lines or by trafficjams. To overcome the latter problems, the volume of traffic on evacuation routesshould be strictly controlled and, in some cases, limited to public transport only.Island communities relying on causeway or ferry links are particularly vulnerable.In extreme cases even 24 hours’ warning may not be sufficient to effect a completehorizontal evacuation, due to either low traffic capacity or distance. It would, forexample, be difficult to effect a total evacuation, even with an adequate stormwarning, in cases where islands are linked to the mainland by long tortuous, low-lying causeways.

Vertical evacuation entails the movement of people into secure storm-proofedhighrise buildings. Such an operation is obviously less cumbersome than horizontalevacuation, but it still requires careful planning. Such planning may be tackled in

3.13.3EVACUATION

3.13.2PREPAREDNESS


two ways; first by designating routes or havens of sufficient capacity to absorb theat-risk population. This may entail road widening, removing or relocating topplinghazards and strengthening building foundations and super structures. Secondly, at-risk communities may impose ceilings on the size of population or developmentinside the known area.

For any effective cyclone mitigation measure public awareness plays the mostimportant role. People in coastal areas must understand the physical nature of thethreat and its disastrous potential (Rakshit, 1987). They should understand the language of the cyclone warnings so that they can respond effectively. To that end,the warnings should be worded in simple language, easily understood by all.Extensive educational programmes should be conducted. Lectures, film shows, seminars, etc. may be arranged in schools and in public theatres. Non-governmentalorganizations (clubs, scientific organizations, etc.) play a very effective role in masseducation. Wide publicity through different media such as radio, TV and newspapers during the pre-cyclone season can remind people to become alert to thepotential calamity. The fishing community is the most vulnerable as they go to seaand remain there for a considerable period. They must carry at least transistor setsto keep them aware of the latest weather situation. It should be stressed that oncethey are caught in an offshore wind associated with a cyclone, they are likely to bedrawn away from the shore and may not be able to escape the cyclone. Many fishermen may also suffer from a false sense of security as they might claim to haveweathered a number of cyclones in their long careers, but there is no guarantee thatthey will be lucky every time.

The mitigation of cyclone disaster requires progress in four areas:(a) Accuracy of warnings;(b) Expeditious dissemination of warnings;(c) Community understanding of the cyclone; and(d) Effective utilization of the warnings by the community.

Typical pamphlets such as “Know your tropical cyclone warning system”, “Whatdo you do when a cyclone strikes?”, “Beware, cyclone are killers”, etc. can beprepared and distributed free of charge. Placards can be displayed in public placesand at roadsides. A list of cyclone shelters and maps showing evacuation routesshould be prominently displayed. Frequent broadcasts/telecasts direct from cyclonewarning centres can convey the correct assessment of a cyclone threat with anauthority and authenticity which can have a far reaching impact on viewers andlisteners.

Persons entrusted with cyclone distress mitigation work should have a higherlevel of cyclone awareness and education. It is therefore essential to have trainingfacilities for these officials and volunteer group leaders at appropriate levels(UNDRO, 1991). In India, for instance, cyclone familiarization meetings/seminarsare held on a routine basis twice a year where state and central government officialsdirectly associated with cyclone distress mitigation activities participate andexchange knowledge and experience.

To alleviate the distress of the people affected by cyclones the following stepsshould be taken in order of urgency:

(a) Medical aid;(b) The prompt disposal of dead bodies and carcasses;(c) Preventive measures against epidemics such as cholera and other waterborne

diseases;(d) Arrangement for the supply of safe drinking water and food;(e) The carrying out of repairs to tanks and other water stores;(f) The mobilization of building materials and their distribution for repair work; and(g) The supply of cattle feed and fodder.

Cyclone disaster management needs to be a long-term, multi-sectorresponsibility which interacts with, and contributes significantly to, nationaldevelopment. It should be intimately concerned with the root causes of communityvulnerability to hazards. To be successful, vulnerability reduction programmes needhigh level coordination and the support of the entire community (Stenchion, 1997).

3.13.4MITIGATION


3.14 OTHER MEASURES

Cyclone-prone areas should be subject to land reform. The harvests should be sophased that they coincide with the safe period or periods of low risk. Better communications are also needed, both to improve the speed of information and toaid evacuation. Protective forests (mangrove) should be encouraged, both to reducethe frequency of inundation and to initiate reclamation, although it is not easy todivert limited resources for such measures.

Risk zone mapping and analysis of land use pattern should be undertaken toguide growth and development away from cyclone-prone areas. Land use legislationand building regulations should be established and strictly enforced in cyclone-proneareas. Architects and planners have a great responsibility in the field of disastermitigation. They must discourage the development of primary social functions,vulnerable production facilities and human settlements in cyclone-prone areas.

All existing public or community buildings such as schools, hospitals, etc.should be made totally safe against cyclones. Easy exits and access to structuresshould also be ensured. Raised platforms for livestock, emergency food grain storagefacilities, drinking water storage and wells with covers to avoid pollution and siltingduring inundation should be built in cyclone-risk areas and properly maintained(Shanmugasundaram, et al., 1993). Action plans for health care after a disastershould be prepared in advance in cyclone-prone areas and implemented if necessary.Such plans should be published, volunteers trained and health kits distributed.Training for survival during a cyclone should form part of such action plans(Madhava Rao, 1990).

A syllabus on topics connected with cyclones and floods may be included inschool and college curricula in disaster-prone areas. The general public and industrymay be educated about insurance schemes against natural disasters and beencouraged to be insured.

3.15 SUMMARY AND CONCLUSION

When a tropical cyclone approaches a country, the threats are threefold – winds,river floods and storm surges. Although strong and violent winds are a fundamentalcharacteristic of tropical cyclones, it is often the rain-induced river floods and thestorm surges which cause the heaviest loss of life and do the greatest agriculturaland other damage. In any planning of disaster prevention and mitigation, the hazards likely to arise from winds, river floods and storm surge should be analyzedseparately and also collectively, particularly in coastal areas.

A tropical cyclone as a single event is a test of the effectiveness of a country’sability in disaster prevention and preparedness. By the time a tropical cyclone hasformed over the ocean and been detected, there will not usually be much time todo more than implement the emergency measures that already exist. Themeteorological service no doubt will make every effort to predict the movement ofthe cyclone and, as it approaches the country, issue warnings to responsible officialsand the general public, but the protection of the country and its people largelydepends on measures that have been taken over several or more years in the recentpast and on the efficiency and zealousness with which they are applied during anemergency.

Cyclones cannot be prevented but an increase in public awareness and effectivepre-cyclone, on-cyclone and post-cyclone measures can definitely reduce theirpotential disaster to a very large extent. It is common knowledge that properattention is never given before the occurrence of calamity; we often forget thatprevention is better than cure. The cooperation of the public during a calamity isbest guaranteed by ensuring that full information and advice is readily madeavailable to all. For this purpose the mass media – radio, television and the press –are extremely important.


3.16 RECOMMENDATIONS

(a) Steps should be taken to encourage the building of capabilities in the areas ofcyclone forecasting and the quick transmission of information to the area likely tobe affected, as well as long-term measures to build up a physical infrastructure thatfacilitates the organization of rescue and relief operations in a systematic manner inorder to minimize loss of life.

(b) Efforts should be made for the setting up of a number of wind observatories, theinstallation of high power storm detection centres, the upgrading of radio stationsto communicate weather warnings and the streamlining of the administrativesystem for effective utilization during rescue and relief operations.

(c) Long-term measures such as (i) construction of cyclone shelters and identificationof existing buildings to serve as cyclone shelters; (ii) afforestation along the coastto reduce wind velocities in order to minimize damage; and (iii) the linking up ofcoastal villages by a reliable road network to facilitate the quicker disbursal of relief.

(d) Cyclone stores should be established in various cyclone-prone districts consisting ofessential drugs, large cooking vessels for use in relief camps, equipment for cleaningdebris and road clogs, pumps, generators, etc.

(e) The coastal villages which are frequently affected by cyclones should be providedwith cyclone shelters which not only withstand high velocity winds but which arealso located at elevated places to provide shelters even during storm surges. Wherethe construction of shelters is not possible because of physical or financial constraints, existing public and private buildings may be identified for providingshelter.

(f) The construction of permanent houses, reconstruction of damaged infrastructure,especially irrigation, drainage and communication systems, inputs to agriculturists,especially for horticultural crops, the removal of sand from fields and the desalina-tion of affected fields, the supply of boats and fishing yarn, repairs to inlandfisheries, etc are to be undertaken.

(g) Historical data about cyclones should be compiled, documented and archived andthe existing database widened. Models for pre-disaster and post-disaster management may be developed by research institutions on a priority basis.Research may also be undertaken into promoting studies/pilot projects for deter-mining the useful effects of afforestation and plantations in cyclone-prone areas.

(h) Training and disaster awareness programmes should be made so that disaster-pronecommunities have a more realistic understanding of the risks to their communityand take more practical measures to save life and property.

(i) Efforts should be made to record storm surge and tide data on different sections ofcoastline for the analysis of surge heights along different sections of the vulnerablecoastline required for designing harbour installations, coastal planning and coastalengineering works.

(j) After the cyclone has passed, the public should be informed about what has happened and what the government is doing to meet the emergency needs of thepeople. The public should also be kept informed of the facilities that are beingmade available and should, at the same time, be advised on the action they shouldtake as families or as individuals.

• ADB, 1992: Disaster management – a disaster manager’s handbook. InformationOffice, Asian Development Bank, P.O. Box 789, 1099 Manila, Philippines.

• Anthes, R.A., 1982: Tropical cyclones: their evolution, structure and effects.Meteor. Monogr., Vol. 19, Amer. Meteor. Soc. Boston, MA, (ISBN 0-033876-54-8), pp. 208.

• Basu, A.N., 1999: Protection of coastal areas from storm and tidal inundation.In: Natural disasters some issues and concerns. Natural Disasters ManagementCell, Visva Bharati Santiniketan, Calcutta.

• Bedson, G., 1997: Specific aspects of natural disasters which affect agricultural production and forests, particularly wildland fires, severe local storms and hurricanes. CAgM Report No. 73, WMO, Geneva.

REFERENCES


• Britten, N.R., 1987: Disaster in the South Pacific: impact of tropical cyclone“Namu” on the Solomon Islands, May, 1986. Disasters 11:120–133.

• Das, P.K., Sinha, M.C. and Balasubramanyam, V., 1974: Storm surges in the Bayof Bengal. Quart. J. Roy. Met. Soc., 100:437–449.

• DHA, 1994: Mozambique, Cyclone “Nadya” Situation Report, No. 5. (information courtesy of R. Gommes – FAO).

• Dvorak, V.F., 1975: Tropical cyclone intensity analysis – forecasting from satellite imagery. Mon. Wea. Rev. 103:420–430.

• Elsberry, R.L. and Marks Jr., F.D., 1999: The hurricane landfall workshop summary. Bull. Amer. Meteor. Soc. 80:683–685.

• ESCAP/WMO/LRCS, 1977: Guidelines for disaster prevention and preparedness intropical cyclone areas. Geneva/Bangkok.

• ESCAP/WMO, 1991: Typhoon Committee Annual Review, WMO, Geneva.• FAO, 1982: Salvage and rehabilitation in the forestry sector of the hurricane affected

areas of the eastern region, Jamaica. FAO, Rome, pp.16.• FAO, 1994: Summary of expert consultation on the coordination and

harmonization of databases and software for agroclimatic applications.29 November–3 December, 1993, Rome, Italy.

• Frank, W.M., 1982: Large scale characteristics of tropical cyclones. Mon. Wea.Rev. 110:572–586.

• Ghosh, S.K., 1977: Prediction of storm surges on the coast of India. Ind. J.Meteor. Geophys. 28:157–168.

• Gommes, R., 1997: An overview (extreme agrometeorological events). CAgMReport No. 73, TD No. 836, WMO, Geneva.

• Gray, W.M., 1968: Global view of the origin of tropical disturbances and storms.Mon. Wea. Rev. 96:669–700.

• Gray, W.M., 1975: Tropical cyclone genesis. Dept. of Atoms. Sci. Paper No. 232,Colorado State University, Ft. Collins, Co., pp. 121.

• Hartman, L.M., David, H. and Giddings, M., 1969: Effects of hurricane stormon agriculture. Water Resources Research 5(3):555–562.

• Holland, G.J. and Elsberry, R.L., 1993: Tropical cyclones as natural hazards: achallenge for the IDNDR. In: Tropical cyclone disasters, Lighthill, J., Emanuel,K., Holland, G.J. and Zhang, Z. (eds.), pp. 17–30.

• Krishna Rao, A.V.R., 1997: Tropical cyclones – synoptic methods of forecasting. Mausam, 48(2):239–256.

• Madhava Rao, A.G., 1990: Cyclone resistant core units. Coarse Wind DisasterMitigation of Structures, SERC, Madras.

• Mandal, G.S., 1993: Tropical cyclones and their warning systems. In: Naturaldisaster reduction, Mishra, G.K. and Mathur, G.C. (eds.), Reliance PublishingHouse, New Delhi, pp. 128–155.

• MIDINRA, 1988: Danos en el sector agropecuario provocados por el huracan“Juana”. Evaluation preliminar. Managua, November 1988, pp. 39.

• Mishra, D.K. and Gupta, G.R., 1976: Estimation of maximum wind speeds intropical cyclones occurring in Indian seas. I.J.M.H.G. 27(3):285–290.

• Oakley, W.J., 1993: A national disaster preparedness service. In: Natural disasters, Merriman, P.A. and Browitt, C.W.A. (eds.), Thomas Telford, London,pp. 270–281.

• OFCM, 1996: National plan for tropical cyclone research and reconnaissance. Officeof the Federal Coordinator for Meteorology, FCM – P25. 159 pp.

• OSRO, 1984: Madagascar, evaluation de la situation agricole a la suite des cyclones.OSRO. TCP/MAG/4404, FAO, Rome, pp. 43.

• OSRO, 1989: Evaluation of the agriculture situation in eastern Caribbean countriesaffected by hurricane “Hugo”. OSRO 03-89-E, FAO, Rome, pp. 63.

• Pisharoty, P.R., 1993: Tropical cyclones. Bharatiya Vidya Bhavan, Bombay 7,India.

• Rakshit, D.K., 1987: Some aspects of community preparedness plan against cyclone.Proceedings of the US-ASIA Conference on Engineering for MitigatingNatural Hazards/Damage, Bangkok.


• Raghavan, S., 1997: Radar observations of tropical cyclones over the IndianSeas. Mausam, 48(2):169–188.

• Ryan, C.J., 1993: Costs and benefits of tropical cyclones, severe thunderstormsand bushfires in Australia. Climatic Change, 25:353–367.

• Shanmugasundaram, J., Appa Rao, T.V.S.R. and Venkateswarlu, B., 1993:Engineering of structures for cyclone disaster mitigation. In: Mishra, G.K. andMathur, G.C. (eds.), Natural disaster reduction, Reliance Publishing House, NewDelhi.

• Smith, R.K., 1993: On the theory of tropical cyclone motion. In: Tropicalcyclone disasters, Lighthill, J., Emanuel, K., Holland, G.J. and Zhang, Z. (eds.),pp. 264–279.

• Stenchion, P., 1997: Cyclone and disaster management. Mausam,48(4)609–620.

• Sugg, A.L., 1968: Beneficial aspects of the tropical cyclone. J. Appl. Meteorol.,7(1):39–43.

• Tamate, S., 1956: Effect of windbreaks on the decrease of salt content in seawind.Proc. of Int. For. Res. Org. 12th Congress, Vol. 1, pp. 47.

• UNDRO, 1991: Mitigating natural disasters. A manual for policy makers and planners. UNDRO, New York.

• WMO, 1983: Human response to tropical cyclone warnings and their content. TCP-11, WMO, Geneva, reprinted 1992.

• WMO, 1990: Tropical cyclone warning systems, Tropical Cyclone Programme,Report No. TCP-26, TD No. 394, WMO, Geneva.

• WMO, 1993: Tropical cyclone programme report No. TCP-31. TD No. 560,WMO, Geneva.

• WMO/UNEP, 1993: The global climate system. World Climate Data andMonitoring Programme, UNEP, WMO, Geneva.

• WMO, 1994: Climate variability, agriculture and forestry. Salinger, M.J., et al., TN No. 196, WMO, Geneva.

• WMO, 1997: Extreme agrometeorological events. CAgM Report No. 73, TD No. 836. WMO, Geneva.


CHAPTER 4

ASSESSING THE ECONOMIC AND SOCIAL IMPACTS OFEXTREME EVENTS ON AGRICULTURE AND THE USE OFMETEOROLOGICAL INFORMATION TO REDUCE ADVERSEIMPACTS(by K.A. Anaman)

4.1 SOCIAL AND ECONOMIC IMPACTS OF EXTREME EVENTSAFFECTING AGRICULTURE

An extreme event generally refers to a relatively rare natural phenomenon whichcould be geophysical, biological or atmospheric and which affects human society interms of human suffering, damage to infrastructure and loss of life that substantiallyexceeds normal expectations. Chapman (1994) classifies natural hazards into threemain groups:

(a) Those hazards originating primarily from the atmosphere and hydrosphere such astropical cyclones, tornados, thunderstorms, floods, droughts, storm surges and possibly dust storms;

(b) Those originating primarily from the lithosphere such as earthquakes, volcanicexplosions, dust storms, mass earth movements including mudslides, landslides andavalanches; and

(c) Those originating primarily from the biosphere such as wildfires, bacteria, viruses,disease-causing agents and other flora and fauna directly affecting human welfare.

Extreme events affecting agriculture considered under the terms of reference ofthe working group include regional droughts, extreme hot and dry weather, frost,flooding, excessive rainfall, tropical cyclones, storm surges, high winds, hailstorms,heat stress and cold injury, forest fires, locust invasions and volcanic aerosols andbelong to all three groups of natural hazards. They are predominantly from the firstgroup of natural hazards noted above, i.e. hazards originating from the atmosphere.However, hazards from the second group, such as volcanic explosions, aresignificantly important, tend to have considerable regional impact and receive greatmass media attention. An example is the 19 September 1994 volcanic explosionsin Rabaul, East New Britain Province of Papua New Guinea, which not onlydestroyed the town of Rabaul but significantly affected cocoa production and otheragricultural activities of the province. The volcanic explosions also led to othernatural hazards such as mudflows, mudfills and flash floods (Blong and McKee,1994). Natural hazards originating primarily from the biosphere also receive massmedia attention such as the Ebola virus outbreak in 1995 in Zaire. Such hazardsdirectly affect agriculture through the displacement and evacuation of farm workersand animals from infested areas.

Impacts from extreme events on agriculture can be positive or negative. While it iseasier to contemplate negative impacts of extreme events such as droughts, tropical cyclones and floods on agriculture, there are several positive impacts orbenefits of extreme events on agriculture. These positive impacts include increasedrainfall to inland areas from tropical cyclones along coastal areas (Ryan, 1993), thefixing of atmospheric nitrogen by thunderstorms, the germination of many nativeplant species as a result of bushfires and the maintenance of the fertility of flood-plain soils due to flooding (Blong, 1992). Chang (1983, 1984) showed that whileHurricane Frederic caused about US$ 1.6 billion of property damage and otherdirect losses in the State of Alabama, it also led to an influx of about US$ 670 millionin government and private sector recovery funds resulting in a US$ 2.5 millionincrease in the municipal funds for the city of Mobile within 12 months. The overall long-term impact of the hurricane was negative to the State of Alabama.

4.1.2ASSESSMENT OF ECONOMIC AND

SOCIAL IMPACTS OF EXTREME

EVENTS

4.1.2.1Types of impacts from extreme

events

4.1.1CLASSIFICATION OF EXTREME

EVENTS AFFECTING AGRICULTURE

AND RURAL SOCIETY

However, certain sectors of the state economy actually benefited. While positiveimpacts can be shown to occur for some extreme events, it is the negative oradverse impacts of extreme events which are more pronounced and do affecthuman society significantly (Joy, 1991; Mitchell and Griffiths, 1993; Sofield, 1993).These negative impacts are often referred to as damages or losses.

Impacts of extreme events can also be classified as either direct or indirect.Direct impacts from extreme events arise from the physical or direct contact of theevents with people, their property and equipment and animals. For example, whenbush fires come into contact with crops and farm buildings or when droughtconditions directly reduce yields of crops and lead to the death of livestock andpeople. Indirect impacts of extreme events are those induced by the events. Indirectimpacts often occur away from the scene of the extreme event or after itsoccurrence. Indirect impacts include evacuation from houses or even permanentdisplacement from an area, disruption to household and leisure activities, loss ofutilities and basic community services, stress-induced sickness and worry and anxietyabout future extreme events, for example, floods or bush fires (Handmer and Smith,1992).

Important indirect impact is normally termed secondary economic impact, ineconomic jargon. The secondary economic impact of an extreme event is derivedfrom the impact on the local or regional economy due to an extreme eventoriginating or initially affecting one or a number of sectors of the economy. A chainreaction impacts over a period of time, say one year, on different sectors of theeconomy. These impacts can be quantified and measured. For example, prolongeddroughts may lead to direct losses such as death of farm animals but may also lead tovarious indirect effects on the incomes of suppliers of inputs to farmers as a result ofthe downturn in the agricultural sector. Secondary economic impact is estimatedusing input-output analysis or variations of this technique such as computablegeneral equilibrium analysis (Jensen and West, 1986; West, 1993).

In addition to impacts of extreme events being classified as positive or negative,direct or indirect, they can also be classified as tangible or intangible. Tangibleimpacts are those which can be easily measured in monetary terms. Tangible benefitsor positive impacts include, for example, the increased rainfall from severe weatherevents, the amount of which can be measured by meteorologists. Tangible losses oradverse impacts are easily quantifiable losses such as damage to farm buildings frombushfires or floods. Here losses can sometimes be recouped if properties are insured.Intangible impacts are often difficult to measure in monetary terms because they arenot purchased or sold in well-defined markets and hence direct market values donot exist. Intangible losses or adverse impacts are sometimes referred to as non-economic or “social costs” in some of the literature (for example, Togola, 1994).Intangible losses include anxiety and fear of future severe events (Oliver, 1988),inconvenience and disruption to farm work and stress-induced ill health and humanfatalities. Intangible benefits or positive impacts of severe events include improvedpreparedness by the population for future severe events. Finally, impacts of extremeevents can be expressed as short term or long term, depending on the duration ofthe after-effects of the events.

The classification of effects of natural hazards as tangible or intangible is a first steptowards evaluating the impacts of these hazards on agriculture. The estimation ofeconomic losses and impact from natural hazards is often imprecise, with doublecounting common if concepts are not clearly defined and applied consistently.Impact assessment from natural hazards should be based on sound economic principles. The evaluation of adverse impacts of extreme events on agriculture andrural society requires delineating the impacts on society in general from those onindividuals, households or businesses. The distinction is necessary to determine financial losses applicable to individuals and societal economic lossesattributed to society as a whole. A single economic unit such as a farm is dealt withvia analysis in changes in net income when financial losses are considered. Marketprices are often used to approximate changes in income imposed by extreme eventson individuals. For societal economic analysis, the analysis is done for all members

4.1.2.2Valuation of impacts of

extreme events

CHAPTER 4 — ASSESSING THE ECONOMIC AND SOCIAL IMPACTS OF EXTREME EVENTS ON AGRICULTURE 53

of a defined society affected by the extreme events. The society can be a town, aregion or a nation. Market values are sometimes adjusted to reflect economic scarcity of resources and inherent market imperfections. Regardless of whether anindividual economic unit or the whole society is being considered, impacts arealways measured as the difference between the occurrence and non-occurrence ofan extreme event.

There are several ways of evaluating losses of assets at the farm/firm/household levelwhich are caused by extreme events. The method depends on whether the asset hasbeen completely destroyed or rendered unusable and whether the asset can be adequately repaired or restored to its former state. In the latter case, the relevantcosts are those required to restore the damaged asset to its former state (before theoccurrence of the extreme event). In the case where the asset has been completelydestroyed, there are several alternative methods for valuing losses. These include:(i) the cost or market price of the asset before it was damaged which ever is lower;(ii) the net selling price of the asset before it was damaged defined as the marketprice minus selling cost; (iii) the cost of the asset minus accumulated depreciation;(iv) current replacement cost of the asset before it was damaged; (v) the incomecapitalization method. The income capitalization method involves deriving the netincome of the asset (before it was damaged) in each year over its remaining lifespan and discounting the future net incomes to the present value by the marketinterest rate less the inflation rate. The composite sum of the yearly present valuesof the net income represents the value of the asset. Under ideal competitive market conditions, all these methods should yield identical market values of the asset.However because of market imperfections, an ideal valuation outcome may notalways be possible. Certain methods are appropriate for certain types of asset andthese are described below.

Current assets defined as those that are expected to be realized within oneaccounting period or one year, such as cash, are valued by their face value. ThusUS$ 500 cash lost during the burning of a farmhouse by a bush fire is simply valuedas a US$ 500 loss. Farm commodities, such as rice stored at a farm to be sold in thefuture, are valued at their current market prices less the expected costs of selling theproduce.

Machinery and working assets are those which are considered to have expectedworking lives of between one and five years. Losses of machinery and working assetsdestroyed by severe events are valued based on the condition of the assets beforethe occurrence of the extreme event. If the machinery and working assets arepartially destroyed then the costs to restore them to their previous states are therelevant losses. However, if these assets have been completely destroyed then thereare a number of methods to value the losses. One method normally used formachinery is the cost of the asset if the destroyed asset was new. For old machineryand farm animals, the loss is the equivalent current market value.

Another option is to value the loss of the destroyed asset as its cost minusaccumulated depreciation up to the time of its destruction. For growing field cropswhich are not yet ready for harvest but have been destroyed by an extreme event,losses are valued as the sum of all costs incurred in producing and maintaining thecrops till the time of their destruction. This value should be compared with themarket value of the growing crops if available and the lower of the two assessmentsused to establish the value of the losses. The income capitalization method iscommonly used to value buildings which are destroyed by severe events. Thismethod is sometimes used for valuing losses due to destroyed farm animals based onexpected income over their remaining lifespan (Anaman and McMeniman, 1990).

Aggregate values of losses are often derived by multiplying the average individualfarm losses by the number of farms affected or by simply estimating the losses foreach farm or firm or household and adding up all the losses of affected businesses orhouseholds. Sometimes insurance payouts to households and businesses are used toderive approximate total losses from severe events for a particular region. Joy(1991) estimated that insurance payouts represented about 25 per cent of total

Valuing aggregate tangible losses

Valuing tangible losses at the level ofthe farm/firm/household


losses from severe weather events in Australia. Total losses were then derived bymultiplying the insurance payouts by a factor of four.

Intangible impacts of extreme events are usually difficult to value in economicterms because they cannot be easily measured in monetary terms. They are likely tobe expressed in non-monetary values, for example, through the use of social indicators such as the number of human deaths and the number of people affectedby stress induced by the extreme event. For example, a summary of fatalities fromsome natural hazards in Australia since European settlement has been compiled bythe Natural Hazards Centre of Macquarie University, Sydney (Coates, 1996).These natural hazards included heatwaves, tropical cyclones, floods, bushfires,lightning strikes and landslides. The number of human fatalities from bushfires overthe period 1827–1991 was estimated as 678, deaths from floods as 2 207 for theperiod 1803–1996, fatalities from lightning strikes as 650 for the period 1803–1992and deaths from tropical cyclones totalled between 1 863 and 2 312 for the1827–1989 period (Coates, 1996). The qualitative listing of general intangibleimpacts affecting people caused by an extreme event without listing the specificnumber of people affected is also common. Briefly, intangible impacts include: (i) human deaths or fatalities; (ii) loss and destruction of personal and culturalmemorabilia and artifacts; (iii) anxiety, stress and induced ill-health; and (iv) inconvenience and disruption of personal and family daily non-business activities.

Loss of human life is difficult to value in monetary terms and is often avoided byprofessional impact assessors. However, there is growing literature developed byeconomists to attempt to put a monetary value on human life (Field, 1994; Dwyer,1986). These methods include the contingent valuation method based onindividuals’ willingness to pay to take measures to reduce the possibility (probability)of death from extreme events. Briefly, a contingent valuation study involvesidentifying and describing the environmental improvement being evaluated such asthe supply of advance warning meteorological information to reduce the possibilityof death from the extreme event (e.g. tropical cyclones and severe floods),identifying and selecting respondents, especially from the target group at risk of theextreme events, using scientific sampling procedures, designing and applying asurvey questionnaire through personal, phone or mail interviews, analyzing theresults statistically and aggregating individual responses to represent an aggregatevalue for the group (Mitchell and Carson, 1989; Bateman, 1994; Field, 1994; Drake,1995). The other popular method of valuing human life is the extension of theincome capitalization method described earlier to value life based on the expectedyearly earnings throughout the rest of the expected lifespan of the deceased person.

Destruction of cultural artifacts and sites is not possible to value in monetaryterms because they are generally considered as irreplaceable. It is also difficult toestablish future incomes from these assets when these assets are not sold or bought inwell-defined markets. Anxiety and stress-related sickness are also difficult to value.However, increases in health-related costs induced by the extreme event mayprovide a proxy for a value of its intangible impact.

4.2 ECONOMIC USE OF METEOROLOGICAL INFORMATION AND SERVICESTO REDUCE ADVERSE IMPACTS OF EXTREME EVENTS ON AGRICULTURE

Data are facts or figures often derived from direct observations. Data tend to be rawpieces of facts and figures. When data are processed or transformed into forms usefulto people they become information. Information is also sometimes defined asknowledge communicated or received concerning some circumstance or eventimplying that information involves both the transformation and communication ofdata from one source to another. Economists define information as data which havebeen processed into forms that are relevant and useful to recipients (Davis andOlson, 1985). The supply of meteorological information and data in forms demandedby customers is the provision of a service. Historical daily rainfall figures of a

4.2.1DATA, INFORMATION AND SERVICES

AS ECONOMIC RESOURCES

Valuing intangible impacts


particular location are data. When these data are transformed into monthly andyearly totals to describe the type of crops that can be grown in a particular location they become information to a prospective farmer. The provision of rainfall data and related information as to the suitability of crops grown by farmersis a service. Scientists use data as inputs into models to generate outputs whichbecome information since those outputs provide utility to the scientists or theirsponsors in explaining or better predicting natural or human-induced events.

The services of national meteorological services are often used as importantinputs in the decision making of many businesses of the national economies andalso sometimes by other countries, for example, the sharing of atmospheric data forairline travel. Meteorological services such as public weather forecasts and warningsare also used by households as consumption goods since the public weather servicesprovide personal convenience to individuals for their day-to-day living. Nationalmeteorological services are often responsible for the collection, storage andarchiving of weather and climate data such as rainfall and temperature data for useby future generations. They also have important roles to play in the freeinternational exchange of meteorological information and other servicesadministered by the World Meteorological Organization.

In the context of the production and use of agrometeorological data, processeddata, information and services, it is essentially a feedback process as is illustrated inFigure 4.1. A national meteorological service produces agrometeorological data andinformation (including agricultural statistics), sometimes in conjunction with othernational and international institutes. The agrometeorological processed data andinformation are sent to farmers through a distribution network which involvesverbal contacts, post, phone and facsimile services. In many countries, the key linkbetween the national meteorological service and farmers is the local agriculturalextension office or possibly research office. Officers in the local agriculturalextension or research office work closely with meteorological officers to furtherprocess meteorological information in forms that can be understood by farmers,especially small-scale farmers. Meteorological information has to be tailored to suitthe production of certain crops and the role of extension officers is important toencourage the effective dissemination of information. Extension officers also providea source for feedback reaction and information from farmers to the nationalmeteorological service, thus allowing it to continually upgrade the quality of itsinformation products.


Figure 4.1Diagrammatic representation ofsupply, demand and distribution

of agrometeorologicalinformation and services to

farmers by a nationalmeteorological Service

Increased agricultural production, reducedcosts of production and increased incomes

Extension offices operated by producers’organization and/or the Ministry of Agriculture.

Offices staffed by extension advisors

Use of services byfarming community

Cooperativefarms Small-scale farms Large farms

Monitoring, observation, collection and storage ofmeteorological data and statistics

Production of processed meteorological data, statistics,weather forecasts and information and climate services

Under some conditions, the national meteorological service can reach somelarge farmers in developed countries directly without the need to pass theinformation through extension offices. This can be done through direct phone orpostal contacts or indirectly through the use of facsimile machines to retrieveweather information produced from designated centres.

Agrometeorological data and information are also useful for other decisionmakers in addition to farmers. Government officials in the Ministry of Agricultureand related bodies such as Ministry of Economic Planning may require informationabout the severity of droughts or the predicted amounts of rainfall for a given seasonin order to plan the importation of food to meet any shortfalls in production. Forexample, the Division of Agrometeorology of the National Institute of Meteorologyand Hydrology of Romania regularly produces and sends agrometeorologicalinformation for use by the President of the Republic, the Senate, the Ministry ofWater, Forestry and Environmental Protection, the Ministry of Agriculture andNourishment, the Academy of Agricultural Sciences and Forestry, the RomanianWater Authority and the mass media such as radio, television and newspapers (Baierand Anaman, 1996).

The economic classification of environmental resources such asagrometeorological data, information and resources is best done using the conceptsof indivisibility and non-exclusiveness (Randall, 1981). A good service is indivisible(or non-rival) if its consumption by an individual does not reduce its availability interms of quantity or quality to others. A good service is non-exclusive (orexcludable) if individuals cannot be excluded or if it is difficult to excludeindividuals from using it. Exclusiveness is achieved through the charging of fees, thesetting up of direct barriers and/or very high entry hurdles and the non-provision ofinformation about availability of the service so that some individuals cannot accessit because of ignorance of its existence. Based on the concepts of indivisibility andnon-exclusiveness, services produced by national meteorological services can beclassified into four main groups:

Indivisible (non-rival) and non-excludable goods such as basic public weatherservices and tropical cyclone warning services which are freely accessible to allindividuals through the mass media. This group also includes weather and climatedata collection and archival services which allow climate data to be made availableto both current and future generations and international public goods such as freely-exchanged meteorological data among member countries of the WMO. Indivisibleand non-exclusive goods are often called “public or collective goods” and are usuallyfinanced by governments (Field, 1994).

Indivisible (non-rival) and excludable goods such as specialized exclusiveservices for certain farmers such as the Cottonfields Weather Service and Weather-by-Fax Services in Australia (Anaman and Lellyett, 1996a). These services requireusers to pay some fees but the use of the services by one individual or firm does notreduce the quality and quantity of the service to others, at least up to a certain pointof congestion.

Divisible (rival) and non-excludable goods such as services dealing with therecording of certain extreme meteorological events where the number of weatherstations recording these events may be declining over time. This means that futuregenerations will not have access to certain types of data available to pastgenerations. In addition, the automation of weather recording using satellites andother technological advances sometimes leads to the non-recording of certain events(phenomena) which are more easily observed and recorded by humans. It is believedin some quarters that satellites might in the future replace human observers forrecording many meteorological events (Myers, 1994). But this might be achievedat a cost of non-recording of several associated weather phenomena which are moreeasily recorded by humans. Automation of the recording of weather events throughtechnological advances appears to contribute in some way to the divisibility ofcertain meteorological services as would be perceived by future generations ofindividuals and businesses.

Divisible (rival) and excludable goods such as specialist services supplied forexclusive use by selected companies which pay commercial market-based rates for


those services. These services are of such a nature that potential users who are notprepared to pay fees for their use can be easily excluded. These services are calledprivate or market goods and are usually produced by private firms. Wheregovernment-financed firms produce these services, consumers often pay the full costof production, including a competitive-based profit margin.

Because of the importance of weather and climate to the production of crops andlivestock, the availability of accurate meteorological information is an importantinput to producers, especially for their weather and climate-sensitive activities.Meteorological information may contribute to the reduction of direct tangiblelosses from extreme events by giving producers signals to act to reduce damage totheir property including crops, livestock and farmhouses, to reschedule weather-sensitive production and leisure activities and also to assure the safety oftheir families. For example, in the context of droughts, meteorological informationmay be used for both operational (tactical) and strategic planning and the management of agricultural activities to minimize the impacts of droughts. Withthe operational use of weather information, small farmers in Mali increased theircrop yields by 20–30 per cent based on a pilot project (World MeteorologicalOrganization, 1997; Togola, 1988). In a strategic sense, weather and climate information may also be used in drought adaptation strategies such as changes inthe seasonal cropping calendar and the use of drought resistant crop varieties.

Information such as weather information is communicated as a signal from thesource of origin (e.g. national weather service) to another point (e.g. homes orfarms) where it is used by producers to plan and manage their business operations.Information is transmitted through a medium such as radio or television. Thetransmission of information signals can be affected by engineering problems whichmay make these signals difficult to understand, especially if the equipment carryingthe signals is outmoded. The transmission may also be affected by problems relatedto whether users are able to understand the messages contained in the informationand make the best use of it.

Meteorologists define the quality of meteorological information in terms oftechnical attributes such as skill, reliability, resolution, sharpness, uncertainty andaccuracy (Murphy, 1991; Lynagh, 1990). From an economic perspective, the qualityof meteorological information is directly related to the degree of usefulness of theinformation to improve decision making by consumers. The economic quality ofmeteorological information may therefore subsume technical measures of quality.Economic quality is often measured by a number of characteristics or attributes.These attributes may be expressed in one or more dimensions, for example, physicalstructure or aesthetic appeal. Some important economic quality attributes ofmeteorological information to users include: (i) relevance (i.e. the extent ofusefulness of the information for a given purpose); (ii) ease of understanding; (iii) aggregation (i.e. the level of detail contained in the signal or message); (iv) accuracy; (v) impartiality; (vi) convenience; and (vii) timeliness (Mitchell andVolking, 1994). These attributes determine the economic value of the information.Since information is interpreted differently by users, quality of information will beperceived differently by users. Key quality attributes are discussed below.

Meteorological information that is relevant meets the needs of users. Users willonly acquire information that is relevant. Relevancy of meteorological informationis its most important attribute. Because of the many uses of meteorological information, national weather services have to produce different types of informationfor different groups of agricultural producers, distributors and consumers.

Meteorological information must be provided in a form that can be understood byend users such as farmers. Such information is often provided in the language oflocal producers and in a format that can be understood and interpreted by users.Meteorological information that is not easily understood will likely be regarded asirrelevant. However, not all information that is easily understood by farmers wouldbe regarded as relevant.

4.2.2.2Ease of understanding

4.2.2.1Relevance

4.2.2DESIRABLE ATTRIBUTES OF

METEOROLOGICAL INFORMATION

AND SERVICES


The level of desired aggregation or detail of useful data contained in an information signal required by users varies from individual to individual and thepurposes the information will be used for. For example, some farmers may sometimes only require forecasts about whether it will rain or not for harvestingdecisions. However, others may at times require information on the expected intensity of rainfall for activities such as irrigation. Decision makers may also incurcosts to enhance the quality of weather information contained in a messagethrough more detailed provision of information about expected weather events.

Perceptions of impartiality of information relate to the degree of trust between theweather information provider and its users regarding whether the informationbeing supplied has the level of quality based on verbal and/or written agreementsbetween the producer and users. Users who trust their information providers therefore assume that the information being received, while not necessarily perfectly accurate, is the best available given the technology, resource constraintsand mutual agreements. Asymmetric market information may exist because agrometeorological information providers (sellers) of weather and climate information tend to know the quality of their products better than the buyers(users). Hence a high degree of trust between sellers and buyers of weather and climate information is necessary for the efficient use of such information.

Convenience is an important attribute of information desired by users. The convenience of information to users involves the ease of access to the informationand also the ease with which the acquired information can be used. Informationwhich users can acquire inexpensively for their decision making processes can beclassified as relatively convenient.

The attribute of timeliness implies that the information is supplied at the timerequired by users. This attribute is indeed very important since information supplied at the wrong time may be worthless. Because the quality of meteorologicalinformation can degrade quickly over time due to rapid changes in conditions ofthe atmosphere, timeliness of such information to agricultural producers oftenimplies having access to the latest available information, for example, rainfallforecasts for the next day, to decide whether to irrigate or not.

Accuracy is a perennial problem in weather forecasting and the provision of meteorological information and services. Accurate information is desired by producers because it reduces the chances of errors in decision making. However,despite all the technological developments of the modern era, weather and climateforecasting is an inexact science and it might never be possible to produce perfectlyaccurate forecasts due to the inherent uncertainty of the natural world. Decisionmakers who use meteorological information are aware of this situation and oftenrequire a reasonable operational level of accuracy from weather informationproviders rather than perfect accuracy. Users of weather and climate informationattach some degree of subjective probability of the occurrence of the forecasted natural event. Sometimes the probability of the event occurring is supplied by theweather information producers. Weather and climate information may have highlevels of attributes such as ease of understanding and convenience and can be readily used to plan weather information-sensitive activities. However, its overallusefulness and consequently its economic value may still be low for many decisionmakers if it is consistently inaccurate. Decision makers could suffer losses if theaccuracy of the information supplied to them deteriorates and they continue to usethat information.

Table 4.1 illustrates the economic quality attributes of two meteorological servicesproduced by the Australian Bureau of Meteorology: (i) a specialist weather information service for cotton farmers in the State of New South Wales called theCottonfields Weather Service; and (ii) the basic public weather service (forecastsand warnings) for the Sydney metropolitan area, also in the State of New South

4.2.2.8Empirical illustration of attributes

of weather information services

4.2.2.7Accuracy

4.2.2.6Timeliness

4.2.2.5Convenience

4.2.2.4Impartiality

4.2.2.3Desired level of aggregation or

detail


Wales (Anaman, et al., 1998). The Cottonfields Weather Service is an enhancedform of the basic public weather service provided to the cotton growing areas ofNew South Wales. The service provides, at some cost, the latest weather information and forecasts for a particular district which are much more detailedthan the basic public weather service provided freely through the mass media. Thisservice is available to producers through facsimile machines. Charges are imposedper minute for retrieving weather information from the service. The basic publicweather service, on the other hand, is available free of charge to householdersthrough the mass media in the Sydney metropolitan area. This is used mainly byurban householders but also by hobby farmers, gardeners and farmers within closeproximity of the metropolitan area and professionals in agricultural marketing and processing.

Ease of understanding the information was ranked as the highest valuedattribute for both services. Not surprisingly, the benefit/cost ratios of the serviceswere high, partly because users understood the messages contained in theinformation. For all the common quality attributes (i.e. ease of understanding,overall usefulness, relevance and accuracy), the Cottonfields Weather Service wasconsidered by users to be higher than the basic public weather service for the Sydneymetropolitan area. This specialist weather service for the cotton industry wasregarded as the better quality product because of the enhancement of the basicpublic weather service for explicit needs and use by cotton growers. The lowestranked attribute for the two services was accuracy. However, accuracy of bothservices was generally considered to be satisfactory with an average ranking above 3.0.

Economic theory attempts to explain the actions of individuals using simplifiedmodels that capture the essentials without all the details observed in the real world.Economic analyses are driven by economic theory and are underpinned by threemain components: human beings, products and resources (McInerney, 1987).Human beings are at the centre of the science of economics. Human beings havedesires for things. These desires generate the driving force for economic activitiesand are satisfied when people get the goods and services they want. These goodsand services are produced using resources or inputs (Dijkhuizen, et al., 1994). Theyare then distributed to people who want them through private and institutionalmarketing systems.

Actions by individuals involve costs and personal sacrifices. These actions oftencreate value. Costs are incurred in producing goods because the resources needed toproduce the goods have to be purchased or supplied by the producer. The relevantcost from an economic perspective is the opportunity cost. The opportunity cost ofa resource is the return that could be realized if the resource is used in its bestalternative use instead of in the production of goods as mentioned earlier. Under

4.2.3INTRODUCTION TO THE ECONOMIC

THEORY OF MARKERS


Table 4.1Users’ quantitative assessment of thequality of two meteorological services

produced by the Australian Bureau ofMeteorology in terms of the averageranking of various economic quality

attributes of information andbenefit/cost ratio of services

Quality attribute of Cottonfields Weather Service Basic public weather service forinformation for cotton growers in use by householders in the

New South Wales* Sydney metropolitan area

Ease of understanding 4.3 (0.12) 4.2 (0.17)Overall usefulness 4.2 (0.14) 3.8 (0.20)Relevance 4.1 (0.15) 4.0 (0.24)Adequate level of details 4.1 (0.15) –Timeliness (provision of information withsufficient time to modify decisions) 3.9 (0.13) –Accuracy 3.6 (0.14) 3.2 (0.25)Frequency (time of day) 3.5 (0.23) –Benefit/cost ratio of service 12:1 4:1

* 5 was used to denote that the information was considered excellent in terms of the specific attribute while 1 meant that information providedwas regarded as very poor in terms of the attribute. The scores 4, 3, 2 indicated good, satisfactory and unsatisfactory assessments respectively.The coefficients of variation of the scores (defined as the standard deviations divided by the means) are in brackets.

ideal competitive conditions, the market price of the resource reflects its true cost orits opportunity cost. The value side of human actions rests on the assumption thatpeople have preferences for goods and services and can express their preference forone over another or one package over another. The economic market value of aservice is what an individual is willing and able to pay for it.

A market is defined as a process by which sellers and buyers determine whatgoods and services they are willing to sell and buy and the terms of the contractsinvolved in the transactions. This process takes place in a geographical locationsuch as a farmers’ market or may not involve a specific geographical location, asoccurs with foreign exchange transactions across several continents (Heyne, 1991).A competitive market involves many buyers and sellers interacting with each other.When the quantity of goods or services sought by buyers equals the quantity offeredfor sale by sellers the market is said to be in equilibrium. An equilibrium is asituation where there is stability. The price of the goods at this point is theequilibrium market price. The actual price of goods is what is observed in the realworld and tends to approximate the equilibrium price because of the continuousinterplay of factors influencing demand and supply to correct imbalances. The twocomponents of the market process are supply and demand and are discussed below.

Economists define demand as the relationship between the price of the goods andthe quantity consumers are willing and able to purchase in a given period assumingall other things are constant. This relationship is often an inverse one, with thequantity sold increasing when the price decreases as illustrated by the linePhDemand in Figure 4.2. A given quantity of goods bought by consumers at a particular price is referred to as the quantity demanded. Although price is an important determinant of the demand, there are other important determinantssuch as taste and preference, the price of other goods and services, the income ofconsumers and sometimes the weather. For example, the demand for ice creamincreases in periods of hot weather. A change in any of these determinants causesan increase or decrease in demand. Based on the graphical presentation, an increasein demand is an increase in the quantity demanded of the goods at each and everyprice and is therefore represented by a shift to the right of the demand curve.Likewise, a decrease in demand leads to a shift of the demand curve to the left.

The other component of the market is supply. Economists define supply as the relationship between the quantity of goods that producers are willing to offer forsale and the price of the goods in a given period of time, assuming everything elseholds constant. This relationship is called the supply curve. Because additional(marginal) costs of production tend to rise with the increasing level of production,the supply curve is generally positively-sloped as shown in Figure 4.2 with DS0 as

4.2.3.2Supply

4.2.3.1Demand


Figure 4.2Change in consumer andproducer benefits with a

parallel shift in the supplycurve due to the use of

improved meteorologicalinformation derived from aspecialist enhanced weather

service produced by a nationalweather bureau

the supply curve of the commodity with the use of unimproved information aboutthe natural world. A given quantity of goods offered for sale at a particular price isreferred to as the quantity supplied. Similar to the concept of demand, there aredeterminants of supply other than the price of the commodity. These other factorsinclude the technology of production, costs of inputs or productive resources usedin the production process, the length of the time period under discussion, weatherand quality of weather information service used by managers of businesses. Achange in one of these determinants will cause an increase or decrease in supply.An increase in supply is an increase in the quantity offered for sale at each andevery price and is represented by a downwards shift of the supply curve as shown inFigures 4.2 and 4.3 with curve DS0 shifting downwards to become curve CS1 basedon the use of meteorological information. A supply curve is relevant to a particularperiod of time and its shape and position tend to depend on the length of theperiod.

The shape of the market demand of goods varies according to the type of goods andmarket. The sensitivity of the quantity demanded as a result of changes in price alsovaries from one product to another and by market. For some products, a smallchange in price leads to a large change in the quantity demanded while the opposite holds true for other goods. The sensitivity or responsiveness of the quantity demanded of a product to changes in its price is measured by the priceelasticity of demand. The price elasticity of demand is the percentage increase inthe quantity demanded as a result of a 1 per cent change in its price. Hence if theprice elasticity of the demand of tomatoes is –0.5 for a country, it means that a 100per cent increase in the price of tomatoes leads to only a 50 per cent decrease inthe quantity of tomatoes demanded. Goods with elasticities that have absolutevalues of between zero but less than unity are classified as price demand inelastic.Goods with elasticities having absolute values greater than unity are classified asprice demand elastic. Many food commodities tend to be price demand inelasticbecause they tend to be basic necessities of life and hence increases in their pricesdo not dampen their consumption by as much as the percentage increase in prices.

Similarly to the price elasticity of demand, the price elasticity of supply is definedas the percentage change in the quantity of a product supplied as a result of a 1 percent change in the price of the product. Thus if a 1 per cent increase in the price oftomatoes results in a 0.5 per cent increase in the quantity of tomatoes supplied in themedium-term period, then the medium-term supply elasticity of tomatoes is 0.5. Themedium-term period is a reasonable period of say one season or one year that allowsfarmers to adjust to the increasing price of the product by increasing the supply tothe market. The supply curve and price elasticity of supply of a product are directlydependent on the time period under discussion (Mansfield, 1994).

For very short time periods, farmers cannot respond quickly to changes in priceunless they have large quantities of the product in store. Hence the price elasticity of

4.2.3.3Elasticities of demand and supply


Figure 4.3Change in producer

benefits based on constantprice resulting from

perfectly elastic demandand the use of improved

meteorological informationfrom a specialist enhancedweather service produced

by a national weatherbureau

supply in the very short term tends to be close to zero. However, in the mediumterm, farmers can respond to an increasing price of goods by cultivating larger areasof the goods. The price elasticities of supply and demand are therefore importantparameters required to understand the benefits accruing to producers and consumersof goods when technological or managerial advances lead to the efficiency ofproduction.

The use of improved meteorological information by producers leads to increasedefficiency if more output of a product is produced with the same amount of existing resources and given technology. In economic terms, improved informationleads to a downwards shift of the supply curve of the commodity. This shift resultsin changes in economic benefits for producers and consumers. The changes in benefits are referred to in common economic jargon as producer and consumer surplus. Consumer surplus is the excess of what consumers would be willing to payfor a product (rather than go without it) over the amount they actually pay basedon its market price. Consumer surplus increases with improved meteorologicalinformation that results in larger quantities of a product being produced, leading tolower market prices. Thus the benefit to consumers of improved meteorologicalinformation is that the product becomes cheaper, allowing more people to purchaselarger quantities of the product. Vice versa, consumer surplus decreases in the presence of supply shocks such as extreme climatic events, for example, drought.Producer surplus or benefits reflect the amounts by which the market price exceedsthe costs of production (net returns). Changes in both producer and consumersurplus allow for assessment of the economic value to the whole society of improvedmeteorological information rather than its value to one producer.

The changes in consumer surplus and producer surplus as a result of the use ofimproved meteorological information by producers of a commodity are illustratedin Figure 4.2, based on using a parallel shift in the supply curve of the product andlinear supply and demand curves and an initial assumption that we are dealing withonly the local economy. Foreign trade considerations will be introduced in a latersection. With the use of traditional knowledge of climate and weather, the equilibrium market price of the product is Po with an amount Qo of the productbought by consumers. The consumer surplus is therefore denoted by the triangulararea PhAPo. The producer surplus is denoted by the triangular area PoAD. With theuse of improved meteorological information, provided by say an institutionalprovider such as a national meteorological service, the increase in production of thecommodity leads to a new equilibrium price denoted as P1.

Consumer surplus increases in size to the area PhBP1 due to the supply shiftcaused by the use of the improved information. The increase in the consumersurplus is therefore made up of rectangle PoAEP1 and triangle AEB (that is areaPhBP1 minus area PhAPo). The producer surplus with the use of improvedinformation is equivalent to the area P1BC. Hence the change in producer surplus isdenoted by area P11BC minus the area PoAD. Since the vertical difference betweenthe two curves is constant, area FGC is equal to area PoAD. Hence producer surplusmeasured as area P1BC minus area PoAD can also be measured as area P1BC minusarea FGC. This gives rise to area P1BGF as the measure of producer surplus (whichis made up of two components, the rectangle P1EGF plus the triangle EBG. Thechange in producer surplus is therefore equal to the producer surplus added as theincrement from Qo (old production level) to Q1 (new production level) and couldbe estimated from either of the two supply curves. The net total economic gain fromthe use of improved information by producers is the sum of the changes in bothconsumer surplus and producer surplus denoted by the shaded area ABCD.

The use of the improved information about the natural world by producers maysometimes have negligible effect on the market price of the commodity, even thoughproduction may have increased. This is represented by the infinitely-elastic demandcurve in Figure 4.3. This implies that there is no change in the consumer surplusbecause the market price of the commodity with the unimproved informationservice is the same as the market price with the improved information service

4.2.4.1Domestic economy considerations

only

4.2.4VALUATION OF METEOROLOGICAL

INFORMATION AND SERVICES BASED

ON BENEFITS ACCRUING TO

PRODUCERS AND CONSUMERS OF

COMMODITIES THAT UTILIZE

METEOROLOGICAL INFORMATION AS

INPUTS IN THEIR PRODUCTION

PROCESSES


(Anaman et al., 1995). The change in producer surplus is therefore the shaded areashown in Figure 4.3 (area ABCD derived from area P1BC minus area P1AD). Theproducer surplus can be estimated by incremental net returns of producers using theimproved information service through producer surveys. While the change inconsumer surplus from the use of improved weather information is likely to bepositive, the extent of the change is not predetermined and depends on the natureof the supply and demand curves of the commodity measured by the elasticities ofsupply and demand. The mathematical formulae used to derive producer surplus andconsumer surplus based on the elasticities of supply and demand are found in severalpapers such as Rose (1980); Alston (1991); Ott, et al. (1995); Alston, et al. (1995)and Anaman and Lellyett (1996b).

Assuming that the supply shift caused by the use of improved meteorologicalinformation is parallel so that the vertical difference between the two curves inFigure 4.2 is constant and equivalent to the unit cost reduction resulting from theuse of the information, the producer and consumer surpluses can be expressed basedon the elasticities of demand and supply and the old price and production levels, Poand Qo. Let us denote Ed, Es and C as the medium term elasticity of demand,medium term elasticity of supply and the actual unit cost reduction from the use ofimproved information respectively. Expressing the unit cost reduction as apercentage of the initial price (K), K = C/P0, the percentage reduction in the priceof the crop due to the supply shift from the use of the improved information (R) isequal to R = (K*Es)/(Es – Ed). The producer surplus, consumer surplus and the totaleconomic surplus can be expressed as follows:CHANGE IN CONSUMER SURPLUS = Po*Qo*R(1 + 0.5*R*Ed)CHANGE IN PRODUCER SURPLUS = Po*Qo*(K-R)(1 + 0.5*R*Ed)CHANGE IN TOTAL ECONOMIC SURPLUS = CHANGE IN CONSUMERSURPLUS + CHANGE IN PRODUCER SURPLUS = Po*Qo*K(1 + 0.5*R*Ed)

Working example for deriving the economic surplus: assume that the use ofimproved meteorological information supplied to farmers by a nationalmeteorological service reduces the cost of growing cotton by US$ 10 per tonne.Without the use of improved information, the price is US$ 1 000 per tonne of rawcotton and the quantity supplied to the market is 300 000 tonnes. Medium-termelasticities of demand and supply (for a period of one year) are –0.4 and 0.1respectively. The change in producer surplus, consumer surplus and the totaleconomic surplus are calculated as follows:K = (10/1000) = 0.01; R = (0.01*0.1)/(0.1 + 0.4) = 0.001/0.5 = 0.002CHANGE IN ANNUAL CONSUMER SURPLUS = (1 000*300 000*0.002)*(1 + 0.5*0.002*0.4) = US$ 600 240CHANGE IN ANNUAL PRODUCER SURPLUS = (1 000*300 000)*(0.01–0.002)*(1 + 0.5*0.002*0.4) = US$ 2 400 960CHANGE IN TOTAL ANNUAL ECONOMIC SURPLUS =(1 000*300 000)*(0.01)*(1 + 0.5*0.002*0.4) = US$ 3 001 200.

The intersection of supply and demand curves in Figure 4.2 denotes the equilibriumor market clearing prices of the goods in the absence of international trade for thegoods. With international trade, the equilibrium price will be above price P0 if thecountry is an exporter of the product and below price P0 if the country is animporter, as is shown in Figure 4.4, in the case of an importing country. In this case,using rice as an example, at the equilibrium price the domestic quantity supplied isQs0 and the quantity demanded is Qd0. The difference between Qd0 and Qs0 is thelevel of imports of the product.

Let us assume that the country imports small enough quantities of the productthat it does not significantly affect the world price of that product. This is likely tobe the situation of a typical small developing country. We also assume that the use ofimproved meteorological information reduces farmers’ unit costs of the productionof rice enabling rice producers to supply a greater quantity at a given price,represented by a rightwards shift of the supply curve as shown in Figure 4.5. Thisleads to a lower level of equilibrium price with price falling from P0 to Pm

*. There

4.2.4.2Incorporation of international

trade into economic analysis


are increased benefits to local consumers since they can buy a greater quantity ofrice at a lower price. Local rice producers are able to sell a larger quantity of thegoods (Qs1) but at a reduced price (Pm*). Hence producers as a whole would benefitonly if the reduced production costs plus the revenue generated from the largerquantity sold offset the reduction in revenues due to the lower price. This can bedetermined based on local supply and demand elasticities and the import elasticitiesof rice.

Figure 4.6 presents the case of a country which exports a crop for which itsproduction is too small to significantly affect the world price of the crop. Thecountry is a price taker. The use of improved meteorological information results inreduced unit costs of production, thereby shifting the supply curve to the right.Production increases from Q1 to Q2 with the extra production (Q2–Q1) exported.This extra export has no effect on the world price of the crop which remains at Pw.The use of improved meteorological information results in increased benefits forproducers measured by the increase in producer surplus of the size ABDE. However,this result does not hold good if many producers in major producing countries alsouse similar improved meteorological information at the same time. Under thisscenario, the world price of the crop could be reduced through overproduction.Producers may then actually lose if the price reduction resulting from theoverproduction is substantial.


Figure 4.4Consumer and producer

benefits based on farmers’use of traditional knowledge

of weather and climate toproduce a good in the casewhere the country imports

some quantities of the good

Figure 4.5Changes in consumer and

producer benefits due to theuse of improved

meteorological informationderived from specialist

enhanced weather service inthe case where the country

imports some quantitiesof the good

Qm = Qd0 – Qs0 = import of good

Qm* = Qd1 – Qs1 = new level of imports

People place values on their preferences and the use of environmental resourcesand amenities. Based on the axiom that every individual valuation of a resourcecounts, the societal economic value of a resource is derived from the valuationsmade by all individuals. However, a distinction is made between values derivedfrom direct, indirect and potential human uses (use values) enjoyed by the individual valuing the resource in question and values based simply on the existence (non-use) of the resource which is independent of human use. The following values are reported in the literature: direct use value, indirect ecologicalfunction use value, option or potential use value, bequest value and existencevalue.

Direct use value is derived when people use or consume environmentalresources, for example, an individual visiting a national park or using timber from atropical forest. Such values are often measured using market values orsurrogate/indirect market values such as the travel cost method. Indirect ecologicalfunction use value is based on benefits or use derived by humans from anenvironmental resource producing or contributing to an ecological stability desiredby humans, for example, the benefits of flood protection derived from a forest(Bateman, 1994) or the benefits derived from oceans acting as sinks for carbondioxide produced by humans. Option use value is linked with the potential use by anindividual (possibly in the future) of an environmental resource rather than thecurrent use of that resource. These three values, direct use value, indirect ecologicalfunction use value and option use value, are considered use values.

Bequest value refers to the value individuals attach to the preservation of thequality or quantity of a resource such as aspects of the global atmosphere orenvironment for use by future generations. The bequest value is considered a non-use value to the valuing individual. Existence value on the other hand is the valueheld by an individual independent of any actual or potential use of anenvironmental resource by himself/herself or any other human beings, living nowor in the future. Existence value arises because some people may put a value on thepreservation of certain species of animals and plants simply because they care forother non-human living things. This concept of existence value was first used ineconomic literature by Krutilla (1967).

Indirect ecological function use, option, bequest and existence values are oftenestimated by the contingent valuation method whereby individuals are asked duringsurveys to indicate the maximum amounts of money they are willing to pay (WTP)for an improved quality of a resource or the minimum amounts of money that theyare willing to accept (WTA) for a reduced quality of the resource. The use of thecontingent valuation method is due to the lack of actual markets for certainenvironmental resources. Therefore researchers attempt to create artificial orhypothetical markets for these resources through surveys to establish their economicvalues (Mitchell and Carson, 1986; Perkins, 1994). The theoretical basis of the

4.2.5ECONOMIC VALUATION OF

METEOROLOGICAL DATA AS

ENVIRONMENTAL RESOURCES

4.2.5.1Introduction to economic

valuation of environmentalresources


Figure 4.6Measurement of the

change in localproducers’ benefits as a

result of the use of anenhanced weather service

for the production of anexport crop

contingent valuation of resources is based on welfare economics principlesexpounded by Hicks (1937). Welfare economics considers the value of animprovement to a non-priced environmental product to an individual as the incomeadjustment made by the individual for the improvement to the extent that his/herutility or satisfaction remains the same as before the improvement.

Meteorological data such as historical rainfall and temperature figures are collected,stored and archived by national meteorological services. Such climatological databecome information only when they are processed into forms useful to recipientssuch as was done by Ussher (1984) who used historical rainfall data, analyzed themand presented them as drought indicators – information useful for farmers and government policy makers. Information derived from these data has economicvalue if it allows decision makers such as businessmen to improve their decisionmaking. In short, data are inputs required to produce an economic product calledinformation. Without data no information can be produced and an information service cannot be provided.

Government meteorological agencies collect and store large amounts of data ofwhich only a tiny fraction is processed into information that is currently useful toindividuals, research organizations and businesses. For example, considerableamounts of climate data stored in government archives have great potential for useby current and future generations, for instance in the planning of future economicdevelopment projects and for research projects in many fields such as climate changeimpact assessment. The economic value of meteorological data therefore includesdirect or current use value and also option use and bequest values. However, becausemeteorological data are not living things, indirect ecological function use value andexistence value do not exist. The three components of the economic value ofmeteorological data are summarized as follows (Anaman, 1996):

Economic Value of meteorological data =Direct Use Value + Option Use Value + Bequest Value

Government policy in many countries often ensures that meteorological dataare made available to users free of charge or at a minimal price based on search andretrieval costs (excluding collection and storage costs). Hence the market price ofsuch data does not reflect the economic value of the data or even the direct usevalue of the data. As with many underpriced environmental goods, indirect pricingmethods such as the contingent valuation technique can be used to establish thetotal economic value involving the three component values described above. Thedirect or current use value of meteorological data can be determined based on theamount of money individuals and organizations in aggregate are currently willing topay to acquire the data for their current uses. The option use value of meteorologicaldata can be determined by the amount of money the current generation ofindividuals is prepared to pay to collect and store meteorological data for thepossibility of use later within their lifetime. The bequest value of meteorologicaldata may be determined by the amount of money the current generation is preparedto pay to collect and store data which will be used by future generations ofindividuals and organizations.

Hence, in a contingent valuation study, the researcher needs to elicit separateresponses to all three component economic values. Expenditure incurred by societyat large on the collection and storage of meteorological data reflects the aggregateeconomic value for these data. These expenditures may be less than optimal levelsnecessary to store and maintain large datasets of meteorological data for futuregenerations because of the inadequate recognition of the option use and bequestvalues of these data, possibly due to the relatively short planning horizons ofgovernments.

A contingent valuation survey to elicit the value of meteorological data shouldinclude both householders and businesses since businesses (such as mining firms)often use meteorological data and may have some interest in the preservation of themeteorological data system for the long-term efficient operation of their firms.Ideally, the contingent valuation survey should be made up of two components.These are: a survey of adult householders and a survey of managers of businesses.

4.2.5.2Valuing meteorological data by the

contingent valuation method


A suggested set of open-ended questions to elicit the value of meteorological data islisted below with an introductory section to explain the questions to respondentsfollowed by the specific questions. The introduction is as follows.

With this question our economic researchers would like to assess the economicvalue of meteorological data continuously gathered, processed and archived or storedby the national meteorological service since the advent of record-keeping of suchdata several centuries ago. A consistent way of finding the economic value of goodsand services such as meteorological data is to establish what most people would bewilling to pay for them in comparison to what they pay for other things they use.This question about the economic value of meteorological data is hypothetical innature, designed to allow our research economists to measure the benefits ofmeteorological data to society. It does not imply any change to current governmentpolicy with regard to financing and access to climate data and other types ofmeteorological data.

Meteorological data such as figures for rainfall, temperature, wind speeds andoccurrences of severe events are continuously collected, processed and archived bythe national meteorological and hydrological service so that they can be accessedby the current generation of users and also by future generations of users (includingyour children and grandchildren). Meteorological data are used by a wide range ofindividuals, businesses and government agencies for many purposes such as thedesign of dams, buildings and structures, the determination of the suitability of landfor the production of commodities in primary industries such as crop and livestockproduction and mining. With this introduction could you please answer thefollowing questions:

1. Assume that you are in a situation whereby the only way you can access meteorological data from the National Meteorological and Hydrological Service isto pay a monthly fee.a. In that situation, what is the maximum amount of money that you would

be willing to pay now per month for access to meteorological data for yourcurrent uses?_____________________dollars per month (measures direct use value)

b. In that situation, what is the maximum amount of money that you would bewilling to pay now per month in order to preserve the meteorological datasystem so as to allow you to have future access to such data for your own futurepurposes within your lifetime?_____________________dollars per month (measures option use value)

2. Further, assume that the only way to preserve the collection, processing and archiving of meteorological data in this country for use by future generations is torequest each adult individual in this country (currently living) to voluntarily pay amonthly fee to the National Meteorological and Hydrological Service to achievethis objective. In that situation, what is the maximum amount of money that youwould be willing to pay per month now in order to allow meteorological data to becollected and stored for use by future generations?

_______________________dollars per month (measures bequest value)The use of open-ended questions in eliciting the WTP of environmental

services has been criticized as providing biased results (Arrow, et. al., 1993).However, other authors such as Bateman, et al. (1994) and Anaman and Lellyett(1996b, 1996c, 1997) have established that the open-ended approach offers validanswers in terms of the link between the WTP and factors such as income andquality of service consistent with the economic theory of demand so long as usersof the services are reasonably conversant with them. In order to avoid possibleproblems related to open-ended questions, a referendum approach is sometimes used.This involves asking respondents a valuation question such as “Are you willing topay US$ 1 per month for access to and use of certain types of meteorological data?”.This is supposed to reflect the real market situation where the consumer isconfronted with a specific price of a product and he/she can decide to purchase itor not at that price. However, because of the possibility of price bargaining,alternatives to the referendum format and open-ended questions have also beendeveloped for eliciting the WTP of meteorological services. One such approach is


the use of payment scales. The payment scales method involves giving respondentsa selection of values or prices for an environmental product or service and askingthem to indicate which price they would be prepared to pay (Donaldson et al.,1997). The range of prices from the payment scales approach may be based onprevious WTP surveys of users in closely-similar situations or based on the currentcosts and prices to derive realistic price ranges.

Payment scale question format:1. Assume that you are in a situation whereby the only way you can access

meteorological data from the National Meteorological and Hydrological Servicewas to pay a monthly fee.a. In that situation, what is the maximum amount of money that you would be

willing to pay per month now for access to meteorological data for your current uses?Put a circle next US$ 0 US$ 5 Other amountto the amount that US$ 1 US$ 6 (please indicate)you are sure you US$ 2 US$ 7 _____________would pay US$ 3 US$ 8

US$ 4 US$ 9b. In that situation, what is the maximum amount of money that you would be

willing to pay per month now in order to preserve the meteorological datasystem so as to allow you to have future access to such data for your own purposes within your life time?Put a circle next US$ 0 US$ 5 Other amountto the amount that US$ 1 US$ 6 (please indicate)you are sure you US$ 2 US$ 7 _____________would pay US$ 3 US$ 8

US$ 4 US$ 92. Further, assume that the only way to preserve the collection, processing and

archiving of meteorological data in this country for use by future generations is torequest each adult in this country (currently living) to voluntarily pay a monthly fee to the National Meteorological and Hydrological Service to achievethis objective. In that situation, what is the maximum amount of money that youwould be willing to pay per month now in order to allow meteorological data to becollected and stored for use by future generations?

Put a circle next US$ 0 US$ 5 Other amountto the amount that US$ 1 US$ 6 (please indicate)you are sure you US$ 2 US$ 7 _____________would pay US$ 3 US$ 8

US$ 4 US$ 9

• Alston, J.M., 1991: Research benefits in a multimarket setting: A review. Review of Marketing and Agricultural Economics, 59(1):23–52.

• Alston, J.M., Norton, G.W. and Pardey, P.G., 1995: Science under scarcity: principles and practice for agricultural research evaluation and priority setting.Cornell University Press, Ithaca, New York, 585 pp.

• Anaman, K.A., Thampapillai, D.J., Henderson-Sellers, A., Noar, P.F. andSullivan, P.J., 1995: Methods for assessing the benefits of meteorological services in Australia. Meteorological Applications, 2(1):17–29.

• Anaman, K.A., 1996: An introductory discussion of cost-benefit analysis applied toclimate change issues. Paper presented at the Users of Climate ChangePredictions Experts’ Workshop, Macquarie University, Sydney, Australia, 31stMay 1995, 28 pp. Published by the Graduate School of the Environment,Macquarie University, Sydney, Australia, ISBN 1 86408 247 X.

• Anaman, K.A. and Lellyett, S.C., 1996a: Producers’ evaluation of an enhancedweather information service for the cotton industry in Australia. MeteorologicalApplications, 3 (2):113–125.

• Anaman, K.A. and Lellyett, S.C., 1996b: Assessment of the benefits of anenhanced weather information service for the cotton industry in Australia.Meteorological Applications, 3(2):127–135.

REFERENCES


• Anaman, K.A. and Lellyett, S.C., 1996c: Contingent valuation of the publicweather service in the Sydney metropolitan area. Economic Papers (EconomicSociety of Australia), 15(3):64–77.

• Anaman, K.A. and Lellyett, S.C., 1997: Evaluation of use and benefits of publicweather and climate services by the mining industry in Queensland. QueenslandGovernment Mining Journal, 98(1147):56–61.

• Anaman, K.A., Lellyett, S.C., Drake, L., Leigh, R.J., Henderson-Sellers, A.,Noar, P.F., Sullivan, P.J. and Thampapillai, D.J., 1998: Benefits of meteorological services: Evidence from recent research in Australia.Meteorological Applications, 5(1), 13 pp.

• Anaman, K.A. and McMeniman, S.L., 1990: Economic evaluation of the performance of purebred and crossbred dairy cattle under on-farm conditions inTropical Northern Australia. Tropical Agriculture, 67(2):105–110.

• Arrow, K., Solow, R., Portney, P.R., Leamer, E.E., Radner, R. and Schuman, H.,1993: Report of the NOAA panel of contingent valuation. Federal Register,Washington, DC, USA, 58(10):4601–14.

• Baier, W. and Anaman, K.A., 1996: Regional Programme for Drought Monitoringin Eastern Europe. Final Report of a WMO Mission to Eastern Europe – Hungary,Bulgaria, Romania and Moldova, 24 June to 25 July 1996, Technical CooperationDepartment, WMO, Geneva, 87 pp.

• Bateman, I., 1994: Research methods for valuing environmental benefits. Economicvaluation of benefits from countryside stewardship. Proceedings of a workshop organized by the Commission of the European Communities DirectorateGeneral for Agriculture, Brussels, 7–8 June 1993.

• Bateman, I., Willis, K. and Garrod, G., 1994: Consistency between contingentvaluation estimates: A comparison of two studies of UK national parks. RegionalStudies, 28:457–474.

• Bateman, I., 1994: Research methods for valuing environmental benefits. Economicvaluation of benefits from countryside stewardship. Proceedings of a workshop organized by the Commission of the European Communities DirectorateGeneral for Agriculture, Brussels, 7–8 June 1993, pp. 47–83.

• Blong, R., 1992: Impact of climate change on severe weather hazards – Australia.Department of the Arts, Sport, the Environment and the Territories, AustralianGovernment Publishing House, Canberra, Australia, 35 pp.

• Blong, R. and McKee, C., 1995: The Rabaul eruption 1994: destruction of a town.Natural Hazards Centre, Macquarie University, Sydney, Australia, 52 pp.

• Chang, S., 1983: Disasters and fiscal policy: hurricane impact on municipal revenue. Urban Affairs Quarterly, 18(4):511–523.

• Chang, S., 1984: Do disaster areas benefit from disasters? Growth and Change,(October): 24–31.

• Chapman, D., 1994: Natural hazards. Oxford University Press, London, 174 pp.• Coates, L., 1996: An overview of fatalities from some natural hazards in Australia.

Paper presented to the Conference on Natural Disaster Reduction, SurfersParadise, Queensland, Australia (29 September to 2 October).

• Davis, G.B. and Olson, M.H., 1985: Management information systems, conceptu-al foundations, structure and development. McGraw Hill, New York, USA.

• Dijkhuizen, A.A., Huirne, R.B.M. and Jalvingh, A.W., 1995: Economic analysis of animal diseases and their control. Preventive Veterinary Medicine,25(2):135–149.

• Donaldson, C., Thomas, R. and Torgerson, D.J., 1997: Validity of open-endedand payment scale approaches to eliciting willingness to pay. Applied Economics,29:79–84.

• Drake, L., 1995: Implications of tropical cyclone and tropical cyclone warning serviceon the income of hotels along the Queensland coast. Climatic Impacts Centre,Contribution Paper No. 95/11, Macquarie University, Sydney, 12 pp.

• Dwyer, L., 1986: Environmental policy and the economic value of human life.Journal of Environmental Management, 22:229–243.

• Field, B.C., 1994: Environmental economics: an introduction. McGraw-Hill Inc.,Sydney, Australia, 482 pp.


• Handmer, J. and Smith D.I., 1992: Cost-effectiveness of flood warnings. Reportprepared for the Australian Bureau of Meteorology by the Centre for Resourceand Environmental Studies, Australian National University, Canberra,Australia, 50 pp.

• Heyne, P., 1991: The economic way of thinking. Sixth edition, MaxwellMacmillan International, Sydney, Australia, 587 pp.

• Hicks, J.R., 1937: Value and capital. First edition, Oxford University Press,London, UK.

• Jensen, R.C. and West, G.R., 1986: Input-output for practitioners: theory andapplications. Australian Regional Developments No. 1, Australian GovernmentPublishing Service, Canberra, Australia, 133 pp.

• Joy, C.S., 1991: The cost of natural disasters in Australia. Paper presented at theClimate Change Impacts and Adaptation Workshop, Climatic Impacts Centre,Macquarie University, Sydney, Australia.

• Krutilla, J.A., 1967: Conservation reconsidered. American Economic Review,57:777–786.

• Lynagh, N., 1990: How accurate are weather forecasts? Weather, 45:345–346.• Mansfield, E., 1994: Applied microeconomics. First edition, W.W. Norton

Company, New York, USA, 684 pp.• McInerney, J.P., 1987: An economist’s approach to estimate disease losses. In:

Disease in livestock: economics and policy. Howe, K.S. and McInerney J.P. (eds.),EUR 11285 EN, Commission of the European Communities, Brussels, pp.35–39.

• Mitchell, R.C., and Carson, R.T., 1989: Using surveys to value public goods: thecontingent valuation method. Resources for the Future, Washington, DC, USA.

• Mitchell, V.-W. and Volking, Y.E., 1993: Analysing the quality of managementinformation: a suggested framework. Management Decision, 31(8):2–19.

• Mitchell, E. and Griffiths, D.J., 1993: The Sydney hailstorm of 18 March 1993.Special Report, Bureau of Meteorology, Melbourne, Australia.

• Murphy, A.H., 1991: Weather forecast assessment: Accuracy or quality?Weather, 46:180.

• Myers, B.L., 1994: Benefits to end users of national hydrological and meteorological services/private sector collaboration. Proceedings of the Conferenceon the Economic Benefits of Meteorological and Hydrological Services, 19–23September,WMO, Geneva.

• Oliver, J., 1989: A survey of public interpretation and opinions in Queensland on thepresent tropical cyclone warning system, Part 1: Case study of Tropical CycloneWinifred, and Part 2: Case study of Tropical Cyclone Charlie. DisasterManagement Studies Centre, Cumberland College of Health Sciences, EastLidcombe, Sydney, New South Wales, 77 pp.

• Ott, S.L., Seitzinger, A.H. and Hueston, W.D., 1995: Measuring the nationaleconomic benefits of reducing livestock mortality. Preventive VeterinaryMedicine, 24(3):203–211.

• Perkins, F., 1994: Practical cost benefit analysis: basic concepts and applications.Macmillan Education Pty. Limited, Melbourne, Australia, 396 pp.

• Randall, A., 1981: Resource economics. Grid Publishing Inc., Columbus, Ohio,USA, 415 pp.

• Rose, R.N., 1980: Supply shifts and research benefits: a comment. AmericanJournal of Agricultural Economics, 62:834–877.

• Ryan, C.J., 1993: Costs and benefits of tropical cyclones, severe thunderstormsand bushfires in Australia. Climatic Change, 25:353–367.

• Sofield, T.H.B., 1993: Cyclones and tourism: the economic impact of visitor cancellations to affected areas and attitudes towards media treatment of cyclones.Department of Tourism, James Cook University, Townsville, Queensland,Australia, 30 pp.

• Togola, Y., 1994: Economic impact of agrometeorological information application in Sahelian agriculture system: an example in Mali. Extended abstractsof papers for presentation at the Conference on the Economic Benefits of


Meteorological and Hydrological Services, 19–23 September, WMO, Geneva,pp. 164–167.

• Ussher, A.K.E., 1984: Rainfall deciles as drought indicators for Ghana.Climatology, Agroclimatology and Water Availability Sub-project, North EastGhana Savannah Research Project, Water Research Resources Institute,Council for Scientific and Industrial Research, Accra, Ghana, 93 pp.

• West G.R., 1993: Input-output for practitioners: user’s guide. Version 7.1,Department of Economics, University of Queensland, Brisbane, Queensland,Australia.

• WMO, 1997: World Climate News. No. 10 (January), WMO, Geneva.


CHAPTER 5

ASSESSING THE IMPACT OF EXTREME WEATHER ANDCLIMATE EVENTS ON AGRICULTURE, WITH PARTICULARREFERENCE TO FLOODING AND HEAVY RAINFALL(by G. Johnson)

This report is divided into two major sections, each focused on agrometeorologicalconditions and impacts associated with flooding and heavy rainfall. The first section is a synopsis of responses to a survey completed by 57 countries,investigating their assessments of extreme agrometeorological event impacts intheir countries. The second section describes, from general to specific, the impactsof flooding and heavy rainfall on agriculture. An appendix is included at the end,presenting the complete tabular results of the survey of countries.

5.1 SURVEY OF COUNTRIES’ ASSESSMENTS OF EXTREME WEATHER ANDCLIMATE IMPACTS, FOCUSING ON FLOODING AND HEAVY RAINFALL

The WMO CAgM Working Group on Agrometeorology of Extreme Events, underthe leadership of Chairman Dr H.P. Das of India, developed a survey of countriesregarding extreme weather and climate events and their agricultural impacts. Thesurvey identified a number of extreme events including drought, desertification,cold/frost, floods and heavy rainfall, high winds and severe storms, tropical storms,forest and range fires and volcanic eruptions.

The survey and the analyzed results provide a significant source of informationrelative to the first two goals of the four terms of reference of the Working Group,which were identified at their meeting in Geneva in April 1997:

(a) Survey and summarize the existing knowledge base regarding the nature andimpacts of extreme weather and climate events on agriculture;

(b) Provide examples of the use of such information from various countries;(c) Help design and establish a database of extreme weather and climate events which

have agricultural impacts, and document these impacts; and(d) Study the requirements for instrumentation that will ensure continuous and

appropriate observations of extreme events.This chapter presents an analysis of the survey and is broadly focused on the

general aspects of extreme event impacts in various countries. There is a specificfocus on flooding and heavy precipitation events, including their impact onagricultural production, their relationship to other extreme events and mitigationstrategies adopted or proposed by various nations in dealing with flooding and/orheavy rainfall.

The survey questions with the yes/no (or missing) responses from each of the 57countries is contained in the appendix as Table A.1. Each of the questions areshown and are numbered on the left side of the spreadsheet. Results are summarizedby question on the last page of the analysis. In addition, and for further analysis,countries were classified into two general and two specific categories: continent andmain climate divisions and sub-continent and specific climate divisions. TheKöppen climate classification scheme (Table 5.2) was adopted for this purpose.Several nations had multiple climate divisions; the major climate division of themost significant agricultural production region of each country was chosen.

A total of 57 countries responded to the survey. In response to the overallquestion, “Does agriculture and/or livestock in your country get affected by one ormore of the following extreme events?”, those events which were reported in themost countries included drought (91 per cent), local severe storms (83 per cent),floods (79 per cent), frost (74 per cent) and high winds (72 per cent). Those events

5.1.1.1Overall survey results for all majorextreme agrometeorological events

5.1.1OVERVIEW OF SURVEY AND

ANALYSES

having agricultural impacts which were reported by the fewest number of countriesincluded volcanic eruptions and earthquakes (30 per cent), locust and grasshopperinvasions (33 per cent), tropical storms (37 per cent), storm surges (37 per cent)and excessive water pollution (39 per cent). Figure 5.1 provides a graphicalrepresentation of these results.

More than 56 per cent of the responding nations had methods of predictingextreme events and nearly 85 per cent had formal warning services. About 68 percent had methods for monitoring extremes as well as routine observations. Only 53per cent of the countries used remote sensing technology for monitoring orprediction. Mitigation strategies were utilized in two-thirds of the countries. Slightlyover one-third of the responding countries (39 per cent) felt they had adequateinstrumentation to fully record and document extreme events, while 21 per centfelt their instrumentation was inadequate. Very few countries (25 per cent) had asystem to assess the socio-economic impact and benefit of extreme events onagriculture, while a large percentage (70 per cent) felt that public awareness trainingfocusing on extremes was important.

A more thorough analysis was then conducted of those 45 countries which reportedagriculturally-related flood problems and the 36 which reported problems associatedwith heavy rainfall. The distinction between flooding problems and heavy rainfallproblems is somewhat diffuse, although their impacts are sometimes quite distinct(for instance, the lodging of mature crops, which can occur with improperly timedand/or intense rainfall, but in the absence of flooding). Of the 45 countriesresponding “yes” to the flood question (No. 1.6), 33 of these also reported problemsassociated with heavy rainfall (73 per cent). Thus, there were 12 countries whichreported flood problems but no problems with heavy rainfall. These included:Chile, the Czech Republic, Ecuador, Egypt, France, Hungary, Ireland, Morocco, theNetherlands, Slovakia, Sudan. Whether it was simply an oversight by the survey-takers in these countries to respond in this way, or whether it truly reflects theconditions in their countries is unclear. In some cases, it is likely that floods occurnot as a direct result of heavy rainfall, or at least not at the site of agricultural production, but rather by transport of water from supply regions. This could certainly be the case in Chile, where there can be significant run-off due to melting snow, and in Egypt, where water is transported down the Nile from areas ofheavy precipitation in central Africa.

An examination of the relationship between flooding and heavy rainfall, andresponses to inquiries regarding certain other extreme events and mitigation/monitoring strategies, was conducted using the survey template results. Specifically,the relationships between responses to questions 1.6 and 1.7 (flooding and heavyrainfall, respectively) on the survey and responses to questions 1.10, 1.11, 4.1, 4,2,

5.1.1.2Analysis of survey results concerning

flooding and heavy rainfall


Figure 5.1Percentage of countries reporting

agricultural impacts from extremeagrometeorological events

Table 5.1Survey results

Question % “Yes” % “Yes”Q.1.6 Q.1.7

(Flooding) (Heavy rainfall)

1.10 68% 56%1.11 33 374.1 46 394.2 56 464.2.2 46 404.3 70 585 51 426 46 396.3 32 257 25 198.4 60 53

Sever

e st

orms

Droug

hts

Floods

Frost

High

winds

Forest

fire

s

Extre

me

heat

Heavy

rainf

all

Extre

me

cold

Deser

tifica

tion

Storm

surg

e

Tropica

l sto

rms

Locu

sts

Volca

noes

4.2.2, 4.3, 5, 6, 6.3, 7 and 8.4 were investigated. Table 5.1 summarizes the responsesof those countries which answered “yes” to either 1.6 or 1.7 and each of the otherquestions. The numbers shown in the table are percentages of the 57 countrieswhich responded to the questionnaire.

In nearly all cases, the joint “yes” responses to each of the questions is similar forboth flooding and heavy rainfall, with approximately 5–10 per cent fewer countriesreporting heavy rainfall problems than flooding problems. The exception is withquestion 1.11, which is tropical storm activity. In that case, there were 4 per centmore countries reporting heavy rainfall and tropical storm damage to agriculturethan there were countries reporting both flood and tropical storm damage.

For those countries that reported flood problems, a much higher percentage alsoreported problems with local severe storms (68 per cent) than those reportingdamage from tropical storms (33 per cent). Similarly, only 20 countries reportedagricultural impacts from storm surges (question 1.8, not shown). Thus, of the 45countries reporting flood damage, a majority of them (25) apparently had noflooding as a result of storm surges; flooding is therefore the result of heavy rain,snowmelt and other factors.

Countries reporting flood problems were more likely to have warning services(question 4.3 – 70 per cent) than they were to have methods of routinelymonitoring extreme events (question 4.2 – 56 per cent) or have predictioncapabilities (question 4.1 – 46 per cent). It is unclear from the survey how 24 percent of these countries could have warning capabilities without havingforecast/prediction services, but more than likely they either purchase or obtainforecast services from other nations.

A relatively high number of counties (60 per cent) felt it was important toconduct public awareness programmes and training on how to cope with and preparefor flooding and heavy rainfall. However, of those countries which indicated thatflooding or heavy rainfall were detrimental to agriculture, a fairly low percentage(25 per cent and 19 per cent, respectively) had any system for assessing the socio-economic impact of extreme events.

Mitigation strategies aimed at reducing the impact of extreme events onagricultural or livestock production were adopted and used in only about half ofthose countries reporting flood problems (51 per cent), and in even fewer countriesreporting damage from heavy rainfall (42 per cent).

CHAPTER 5 — THE IMPACT OF EXTREME WEATHER AND CLIMATE EVENTS ON AGRICULTURE – FLOODING AND HEAVY RAINFALL 75

Table 5.2Abbreviations for continent and

Köppen climate classifications

Abbreviation Continent (CC) Abbreviation Main Climate Division (MCD)

NA North America A Moist tropicalSA South America B Dry climatesE Europe C Moist climates with mild wintersAS Asia D Moist climates with severe wintersAF Africa E Polar climates*AU Australia, South-west Pacific H Highland climates*

Sub-continents (SC) Specific Climate Division (SCD)

NNA Northern North America AF Tropical rain forestCA Central America AM Tropical monsoonNSA Northern South America AW Tropical wet and drySSA Southern South America BW Arid desertNE Northern Europe BS Semi-arid or steppeSE Southern Europe CFA Humid subtropical*ME Middle East CFB or CFC Marine*NAF Northern Africa CS MediterraneanSAF Southern Africa CW Dry winterEE Eastern Europe DFA or DFB Humid continental*SWA South-west Asia DFC or DFD Subpolar*SEA South-east Asia DW Dry winter*EA Eastern Asia ET Polar tundra*AU Australia, South-west Pacific EF Polar ice cap*

H Highland*

Four classification criteria were used to categorize the various countries respondingto the survey, along with their abbreviations or key codes. These four criteria andtheir respective components are shown in Table 5.2. In total, six continents, 14sub-continents, six main climate divisions and 15 specific climate divisions wereused for classification purposes (there were eight specific climate divisions intowhich none of the 57 responding nations fell – these are asterisked).

Table 5.3 provides a list of countries and basic statistics regarding therelationship between these country classifications and reported flood-related andheavy rainfall-related agricultural damage. In general, there were few, and mostlikely statistically insignificant, differences noted in any of the four majorclassifications. This analysis suffered to some degree because of the small number ofcountries that fell into some categories, rendering the reported numbers somewhatsuspicious, at best. Nevertheless, flooding appears to have significant agriculturalimplications in most countries, regardless of geographic or climatic characteristics.Two exceptions would perhaps be northern Europe (code NE) and the Middle East(ME) in the sub-continent categorization. Only 60 per cent of the countries in bothof these regions reported flooding as a significant agricultural problem. Also, only60 per cent of the five countries in the arid desert (BW) category reported floodproblems and in each case it is likely that the flooding occurs either by thetransporting of water into these countries from other regions, or it occurs in aportion of the country that falls into another specific climate classification.

Relative differences between the various classifications were more stronglynoted in the analysis of countries reporting agricultural damage due to heavyrainfall. For example, only 50 per cent of the countries in Africa and South Americaand only 41 per cent of the countries in Europe listed heavy rainfall as a significantagricultural problem. Meanwhile, 83 per cent of the 18 countries reporting fromAsia (a more statistically significant number) reported heavy rainfall impacts. Sub-continent categorization focused this further, with only 20 per cent of the 10countries in northern Europe reporting heavy rainfall problems, only 33 per centfrom southern South America and just 40 per cent of the countries in the MiddleEast and northern Africa. In contrast, 100 per cent of all countries responding fromAustralia and the south Pacific, Central America, East Asia, northern North

5.1.1.3Analysis of countries reporting

flood and heavy rainfall problemswith respect to four classification

criteria


Table 5.3aContinent and floods

Country 1.6 CC

Egypt y AFMorocco y AFNigeria y AFSudan y AFSwaziland y AFChad y AF

No. of countries 6Per cent of the 8 75%

Country 1.6 CC

Argentina y SAChile y SAColombia y SAEcuador y SAPeru y SA


Country 1.6 CC

Belize y NACanada y NAUnited States y NA


Country 1.6 CC

Armenia y ASAzerbaijan y ASBangladesh y ASIndia y ASIran y ASJapan y ASKazakstan y ASMalaysia y ASMyanmar y ASPhilippines y ASRepublic of Korea y ASSri Lanka y ASThailand y ASTurkey y AS


Country 1.6 CC

Australia y AUFiji y AUSolomon Islands y AU


Country 1.6 CC

Austria y ECzech Republic y EFrance y EHungary y EIreland y EItaly y EMoldova y ENetherlands y EPortugal y ERomania y ESlovakia y EUkraine y EUnited Kingdom y E


America, southern Africa and South-east Asia listed heavy rainfall as a problem.Classification at the sub-continent scale was clearly important for identifying centresof action; 67 per cent of southern Europe listed heavy rainfall as significant – some47 per cent greater than northern Europe.

The 57 countries in the survey fell into one of four major climate divisions andthere were noted differences between these divisions as well. Only 48 per cent ofthe countries which had their major agricultural production areas classified as moistclimates with mild winters (Köppen code C) reported problems with heavy rainfall.Meanwhile, 88 per cent of the 16 countries which were classified as moist tropical(A) had heavy rainfall problems, as might be expected. Sixty-seven per cent ofcountries classified with either dry (B) or moist with severe winter (D) climateslisted heavy rainfall as an agricultural problem.

One hundred per cent of all countries with specific climate divisionclassifications of AF (tropical rain forest), AM (tropical monsoon) or CW (drywinter, but moist overall climate) reported heavy rainfall problems. Only 50 per centof countries with a CS (Mediterranean), and 60 per cent of countries with a BWclassification listed heavy rainfall.


Country 1.6 SC

Japan y EARepublic of Korea y EA


Country 1.6 SC

Argentina y SSAChile y SSA


Country 1.6 SC



Country 1.6 SC

Moldova y EERomania y EESlovakia y EEUkraine y EE


Country 1.6 SC

Colombia y NSAEcuador y NSAPeru y NSA


Country 1.6 SC

France y SEItaly y SEPortugal y SE


Country 1.6 SC

Egypt y MEIran y METurkey y ME


Country 1.6 SC

Malaysia y SEAMyanmar y SEAPhilippines y SEAThailand y SEA


Country 1.6 SC

Morocco y NAFNigeria y NAFSudan y NAFChad y NAF


Country 1.6 SC

Austria y NECzech Republic y NEHungary y NEIreland y NENetherlands y NEUnited Kingdom y NE


Country 1.6 SC

Armenia y SWAAzerbaijan y SWABangladesh y SWAIndia y SWAKazakhstan y SWASri Lanka y SWA


Country 1.6 SC

Canada y NNAUnited States y NNA


Country 1.6 SC

Belize y CA


Country 1.6 SC

Swaziland y SAF


Table 5.3aSub-continent and floods

Extreme agrometeorological events are pervasive worldwide. The most significantproblems arise from either too little or too much water. In many cases, with sufficient water, many temperature-related problems can be ameliorated, althoughnot always. Severe local storms are a significant problem in most areas of the worldand their agricultural impacts are often due to their combined forces (e.g. rainfallplus wind).

5.1.1.4Survey conclusions


Country 1.6 MCD

Bangladesh y ABelize y AColombia y AEcuador y AFiji y AIndia y AMalaysia y AMyanmar y ANigeria y APhilippines y ASolomon Islands y ASri Lanka y AThailand y A


Country 1.6 MCD

Argentina y BAzerbaijan y BEgypt y BIran y BKazakhstan y BPeru y BSudan y BSwaziland y BChad y B


Country 1.6 MCD

Bangladesh y AWBelize y AWColombia y AWEcuador y AWIndia y AWNigeria y AWSri Lanka y AWThailand y AW


Country 1.6 MCD

Canada y DUkraine y D


Country 1.6 SCD

Myanmar y AMPhilippines y AM


Country 1.6 SCD

Fiji y AFMalaysia y AFSolomon Islands y AF


Country 1.6 SCD

Fiji y BSMalaysia y BSSolomon Islands y BSFiji y BSMalaysia y BSSolomon Islands y BS


Country 1.6

Republic of Korea y


Country 1.6

Chile yItaly yMorocco yPortugal yTurkey y


Country 1.6 SCD

Egypt y BWKazakhstan y BWPeru y BW


Country 1.6 MCD

Armenia y CAustralia y CAustria y CChile y CCzech Republic y CFrance y CHungary y CIreland y CItaly y CJapan y CMoldova y CMorocco y CNetherlands y CPortugal y CRepublic of Korea y CRomania y CSlovakia y CTurkey y CUnited Kingdom y CUnited States y C


Table 5.3aMain climate division and floods

Table 5.3aSpecific climate division and floods


Country 1.7 CC

Madagascar y AFNigeria y AFSwaziland y AFChad y AF


Country 1.7 CC

Barbados y NABelize y NACanada y NAUnited States y NA


Country 1.7 CC

Austria y EItaly y EMoldova y EPortugal y ERomania y EUkraine y EUnited Kingdom y E


Country 1.7 CC



Country 1.7 CC

Argentina y SAColombia y SAPeru y SA


Country 1.7 SC



Country 1.7 SC

Malaysia y SEAMyanmar y SEAPhilippines y SEAThailand y SEA


Country 1.7 SC

Armenia y SWAAzerbaijan y SWABangladesh y SWAIndia y SWAKazakhstan y SWASri Lanka y SWA


Country 1.7 SC

Japan y EAMongolia y EARepublic of Korea y EA


Country 1.7 SC

Moldova y EERomania y EEUkraine y EE


Country 1.7 SC

Barbados y CABelize y CA


Country 1.7 SC

Iran y METurkey y ME


Country 1.7 SC

Nigeria y NAFChad y NAF


Country 1.7 SC

Austria y NEUnited Kingdom y NE


Country 1.7 SC

Canada y NNAUnited States y NNA


Country 1.7 SC

Colombia y NSAPeru y NSA


Country 1.7 SC

Madagascar y SAFSwaziland y SAF


Country 1.7 SC

Italy y SEPortugal y SE


Country 1.7 SC

Argentina y SSA


Country 1.7 CC

Armenia y ASAzerbaijan y ASBangladesh y ASMalaysia y ASMongolia y ASMyanmar y ASPhilippines y ASRepublic of Korea y ASSri Lanka y ASThailand y ASTurkey y AS


Table 5.3bContinent and heavy rainfall

Table 5.3bSub-continent and heavy rainfall

Survey results found that while the number of countries falling into anyparticular category typically was not statistically significant, nevertheless some usefulinformation about susceptibility to either flooding or heavy rainfall damage and therelationship to various classifications was gained. A similar approach could be usedto discern regional and climatic differences in the susceptibility of agricultural andlivestock production to other extreme events, including wind, frost and drought. Inthis case, it is evident that flooding and/or heavy rainfall are serious problemsplaguing agricultural production in many parts of the world.


Country 1.7 MCD

Bangladesh y ABarbados y ABelize y AColombia y AFiji y AIndia y AMadagascar y AMalaysia y AMyanmar y ANigeria y APhilippines y ASolomon Islands y ASri Lanka y AThailand y A


Country 1.7 MCD

Armenia y CAustralia y CAustria y CItaly y CJapan y CMoldova y CPortugal y CRepublic of Korea y CRomania y CTurkey y CUnited Kingdom y CUnited States y C


Country 1.7 MCD

Canada y DUkraine y D


Country 1.7 MCD

Argentina y BAzerbaijan y BIran y BKazakhstan y BMongolia y BPeru y BSwaziland y BChad y B


Table 5.3bMain climate division and floods

Table 5.3bSpecific climate division and heavy rainfall

Country 1.7 SCD

Myanmar y AMPhilippines y AM


Country 1.7 SCD

Republic of Korea y CW


Country 1.7 SCD

Fiji y AFMalaysia y AFSolomon Islands y AF


Country 1.7 SCD

Argentina y BSAzerbaijan y BSIran y BSSwaziland y BSChad y BS


Country 1.7 SCD

Kazakhstan y BWMongolia y BWPeru y BW


Country 1.7 SCD

Italy y CSPortugal y CSTurkey y CS


Country 1.7 SCD

Bangladesh y AWBarbados y AWBelize y AWColombia y AWIndia y AWMadagascar y AWNigeria y AWSri Lanka y AWThailand y AW


5.2 THE IMPACT OF FLOODING AND HEAVY RAINFALL ONAGRICULTURE

In 1995 the World Meteorological Organization (WMO) commissioned a WorkingGroup on Agrometeorology Related to Extreme Events. This Working Group wascharged with investigating the nature and magnitude of the impacts of extremeweather and climate events on agriculture and with evaluating the various preparations and monitoring strategies for, and responses to, these events in nationsthroughout the world. This has included reviews of literature on the subject, surveys of various national problems and responses and the development of possible ways of mitigating the detrimental effects of extreme events on agricultural production.

Two of the principal hydrometeorological events that often have deleteriouseffects on agriculture are flooding and heavy rainfall. It is recognized that in somecases these events can have positive effects. Examples include heavy rainfall from atropical storm ending a long agricultural drought, or recurring, annual flooding thatreplenishes topsoil and soil nutrients, as was the historic situation along the lowerstretches of both the Nile and Ganges rivers. However, it is extreme events thatdamage agriculture and agricultural production that is the focus of this discussion.

First, a working definition of an extreme agrometeorological event is needed. In theintroduction to Extreme agrometeorological events (WMO CAgM Report No. 73,1997), these events are described as being “at the interface between a vulnerableagricultural system and extreme weather conditions”. Susman et al. (1983) similarly defined a disaster as “the interface between an extreme physical event anda vulnerable human population”. The WMO report further states that “the definition of extreme agrometeorological events is broader, as they include as wellweather conditions conducive to the development of agents (such as pests and diseases) that negatively affect agriculture (the term, according to FAO definition,includes crop agriculture, livestock and pastures, forests and fisheries, both oceanand inland fisheries)”.

Webster’s New Riverside University Dictionary of English Language definesflood as “an overflowing of water onto normally dry land” (noun form), or “anabundant flow or outpouring; to become inundated or submerged” (verb forms). TheAmerican Meteorological Society’s Glossary of Meteorology (1970) defines a floodas “the condition that occurs when water overflows the natural or artificial confinesof a stream or other body of water, or accumulates by drainage over low-lying areas”.

These definitions give the broad, generalized scope of flooding. Floods also havetemporal characteristics; those that occur in a short period of time or come quicklyare called flash floods and are defined by the glossary to be, “Floods that rise and fallquite rapidly with little or no advance warning, usually as the result of intenserainfall over a relatively small area. Other possible causes are ice jams, dam failure,etc.”.

Floods are thus excessive water and in this report the focus is on submergedagricultural areas that normally (in time and space) are not flooded. They may becaused by excessive or heavy rainfall, but this is not a necessary prerequisite for aflood. The glossary definition of flash flooding mentioned ice jams and dam failuresas two possible sources of flash flooding that can occur in the absence ofprecipitation. Other, non-rainfall-induced causes of flooding include rapid snowmeltdue to warm temperatures and/or relatively high dewpoints and the release of waterfrom an upstream impounding structure.

In most situations, however, flooding does not occur unless extremely heavyrain falls. This may or may not be at the location of the flood. We thus define heavyrainfall to be abnormally large amounts of liquid precipitation, in time and/or space.Heavy rainfall is often considered in a point spatial context, as measured by a singlerain-gauge. However, it is important to include both spatial and temporal aspects ofa rainfall event, and its impact, to properly qualify it as a heavy rainfall-inducedextreme agrometeorological event.

5.2.1.1Definitions and background

5.2.1OVERVIEW


The interaction between spatial and temporal scales is important indetermining impacts. Typically, as the temporal characteristics of a rainfall eventdecrease, the potential for significant agricultural impacts decreases as well, withoutan accompanying increase in spatial dimensions. However, significant and evencatastrophic, local damage can occur from a very intense rain storm that covers avery small area. This usually occurs in very intense convective storms that havenegligible movement or move laterally down a watershed. In general, as the spatialdimensions of a storm increase, the amount of precipitation per unit area decreases.Similarly, a large river basin is relatively insensitive to small, isolated storms but islargely affected by weather systems that cover a major portion of the region, like amonsoon depression.

Looking only at temporal characteristics of storms and their consequent impacton agriculture, the concept of intensity-duration-frequency (idf) becomes important.If any of these three storm characteristics increases, the potential for agriculturalimpacts increases. Heavy rainfall at a location can thus be defined using the idfconvention and is framed by the agroclimatology of a location or region in question.Very intense (extreme) rainfall can result in catastrophic flood damage even thoughit occurred for a relatively short period of time, and/or at the proper location, at theproper time of year, etc.

The same intensity of rainfall (given in mm/hr, say) resulting in the sameamount of total storm precipitation (amount = intensity x duration) can haveremarkably different results given a number of factors, all comprising the location’sagroclimatology. In general, greater direct damage to agriculture occurs from stormshaving higher intensities with sufficient duration, compared with low intensity, longduration storms.

For the purposes of this discussion relative to agriculture and extremes, heavyrainfall must include not only the strict meteorological definition (idf), but whatimpact it has. One important factor determining the impact is antecedentconditions. If soil conditions have been quite dry due to limited prior precipitationand/or lack of irrigation, the potential for water absorption becomes greater,resulting in less run-off and potential flood problems. However, if prior rains, orfrozen soils, or other factors limit infiltration, then the same heavy rainfall eventcan cause significant damage.

Other factors include the soil, vegetation and terrain characteristics of thelocation. Bare, hard, sloping surfaces reduce infiltration and therefore increase run-off and potential erosion, on that surface as well as downstream. Accumulating,low-lying areas can be high risk zones for flooding. However, here it is important todiscuss the frequency issue, since these zones may normally be wet (i.e. wetlands)and flooding is neither unexpected nor unwelcome and damaging.

Economic and political factors increase the vulnerability of agricultural landsto extreme weather events, as well. Due to population pressures and other factors,there is a move towards using more marginal lands for agriculture in somedeveloping countries, for example. These lands may be much more susceptible toflooding damage than traditional farmlands nearby.

Thus, relative to flooding and heavy rainfall, it must again be emphasized thatany hydrometeorological event needs to be put into the full agricultural context totruly be considered an extreme agrometeorological event.

Flooding and/or heavy rainfall often has significant, deleterious effects on agricultural production. The impacts can be wide ranging, both temporally, spatially, economically and even culturally and politically. The severity of theirimpacts is often a function of many factors.

Effects of these phenomena on agriculture can be classified as direct or indirect(Gommes and Negre, 1992). Direct effects are those that affect the property andincome of individuals, enterprises and the public sector. An example would be theloss of a current crop of maize due to severe flooding. Indirect effects are slower andoften more widespread (geographically, economically, etc.) than direct effects andresult from decreased income, environmental degradation, and other factors. In

5.2.2CHARACTERISTICS OF FLOODING

AND/OR HEAVY RAINFALL AS

EXTREME AGROMETEOROLOGICAL

EVENTS


general, direct effects are much easier to quantify, while estimates of indirect effectsare often incomplete due to their complicated impacts throughout society.

With this in mind, the following lists give a sampling of the numerous examplesof both direct and indirect effects that flooding and heavy rainfall can have onagriculture. The lists are by no means exhaustive, but represent some of the moresignificant problems that can result:

Non-growing season or fallow period:Loss of topsoilLoss of soil nutrientsSoil compactionSoil erosionDeposition of undesirable materialsPermanent damage to perennial crops, trees, livestock, buildings andmachineryDisplacement of personsBreakage of levees and other retention structuresAnaerobic processesPermanent cessation of farming in floodplainsPermanent diversion/realignment of rivers, streams, other bodies of water andsettlementsLoss of livestock and/or habitat

Growing season:Waterlogging of cropsLodging of standing cropsLoss of soil nutrientsLoss of pasture useSoil erosionGreater susceptibility to diseases and insectsInterruptions to tillage, planting, crop management, harvestingPermanent damage to perennial crops, trees, livestock, buildings andmachinerySoil temperature reduction and/or retardationNecessity of installation of expensive drainage systemsLoss of livestock and/or habitatTransportation interruptionsGrain spoilage, in-field and off-siteFeedback effect, enhancing precipitation due to large, free-water evaporativesurfaces

The prevailing agriculture in a region is largely determined by the climate. Thisincludes temperature (winter minimum, summer means and maximums, ranges,variability), precipitation (total amounts, temporal distribution, extremes, variability), wind, solar radiation, humidity, and other factors. Sustainable agricultural systems are designed to function effectively a high percentage of thetime so that production losses are relatively rare, or never occur. Thus, crops likerice that can function effectively in saturated and even submerged conditions areappropriate for locations that flood regularly and the system becomes dependentupon regular flooding. Many other crops (e.g. corn) would not be adaptable to suchconditions and would not be appropriate alternatives to rice.

Extreme flooding events – those that occur outside anticipated averageconditions and the typical variability about the mean – can cause varying amountsof agricultural damage. The damage can be direct, as to standing crops, or indirect,such as long-term changes to the landscape and soil.

Direct damage to growing plants is most often caused by depletion of oxygenavailable to the plant root zones. Flooding creates anaerobic soil conditions that canhave significant impacts on vegetation. Root and shoot asphyxia, if prolonged,typically leads to plant death. Chemical reactions in anaerobic soils lead to a

5.2.2.1Mechanisms of flood damage;

some examples


reduction in nitrate and the formation of nitrogen gas. This denitrification can be asignificant cause of loss of plant vigour and growth following flooding (Foth, 1978).In the extreme case of regularly flooded areas, like rice paddy fields, or long durationfloods, harmful chemicals like ammonium, hydrogen sulphide and methane developand can build up to toxic levels, requiring special (and sometimes expensive) actionto insure their timely removal from the soil.

Flood severity should be measured by both its direct and indirect effects. Oneimportant determinant of severity is flood duration. In a study by Ritter and Beer(1969) in which they surface-flooded an area of poorly fertilized corn with severalcentimetres of water when the corn was 76 cm high, yield reductions were 14 percent after 24 hours of continuous flooding, but increased to 30 per cent after 96hours of flooding. For fields with much better fertilization, the decrease wassignificantly less (only 16 per cent after 96 hours). During severe floods, nutrientsare often flushed rapidly from the soil, though, so both factors act to createsignificant damage (see section on fertilizing after the California floods – page 93 –for reference).

The timing of a flood is critical, too, for its impact. Floods and excessive soilmoisture during spring planting, germination and establishment periods are typicallymore damaging to final crop yields than is drought during this same period (Raperand Kramer, 1983). Floods during this period result in soil temperature retardationand/or seed destruction. Even in cases where flooding is used intentionally, as forfrost protection of some horticultural crops, the drying-out period afterwards canseverely depress soil temperatures and can result in as much or more damage thanfrom frost alone (Lowry, 1972). As plants become larger and more established, minorflooding becomes less damaging than severe water deficits, especially during criticalpollination periods (Shaw, 1977). Extreme flooding can be catastrophic during thistime as well, though, as evidenced by the Midwest USA flood of summer 1993 (seepage 85).

Preconditioning of an area is very important for determining how significantand damaging a flood will be. Important considerations here are soil, vegetation andwater supply factors. Soils that are saturated prior to an extreme weather event willmore likely result in a damaging flood than soils that are relatively dry. Fields thathave recently been tilled and are devoid of vegetation are much more susceptibleto soil erosion. Vegetation that is able to use much of the water and that can act asa barrier to moving water (horizontally and vertically) can reduce flood severity andimpacts. Water storage systems (rivers, lakes, reservoirs, etc.) that are able to captureand hold most of the incoming water will be effective in reducing flood damage.Thus, water supply managers in snow-fed water supply regions of the world typicallydraw down reservoirs as much as possible prior to the normal beginning of thesnowmelt run-off season. In rain-fed agricultural systems, managers typicallyanticipate rainfall during the growing season sufficient to naturally or artificiallyirrigate crops. In both situations, however, there is often a balance needed betweenretaining enough water for agricultural production and environmental health andmaintaining enough available storage volume to capture incoming water andprevent floods. Here, analyses of past weather and water data are critical forestimating average conditions and inherent variability.

Significant floods in temperate to subpolar regions often occur in the winterand spring seasons when warm temperatures and moderate to heavy rainfall followsa period of cold and/or snowy weather (see California, 1997 example below). Heavysnows can contain significant snow water equivalent (potential liquid water) which,if melted rapidly, can cause rapid and voluminous runoff. A further complicatingfactor is frozen soils which can act as a barrier, much like an impervious subsoil layeror pavement, to water. Instead of infiltrating, this rain or rain-and-snow-fed water isforced to run off, often rapidly reaching main stream channels. The Pacific north-western USA and portions of Europe are two regions where this phenomenon hascreated the most significant and widespread flooding events.

There are several major problems that may reduce agricultural production in aregion for many years following a flood. For instance, if sea water or salty lakes flooda region, the soils may become salinized for many years to come and a significant


effort may be needed to effectively flush these salts from the soil. There may also besignificant siltation and deposition of large quantities of less desirable topsoils. Inthe Midwest USA flood of 1993, as much as 150 cm of sand and other soil wasdeposited on top of rich topsoil in the region near the confluence of the Missouriand Mississippi rivers. Significant effort and major economic cost was needed toremove this material from a highly productive agricultural region, albeit one thatwas created by dyking long stretches of major rivers.

The destruction of protective dykes is often the cause of catastrophic flooding ofagricultural zones, with numerous examples worldwide (Bangladesh; California andMississippi, Missouri and Red River regions of the USA; China). Dyking is oftendesirable, in spite of the typically large expense, because the resulting farmland isextremely productive. However, the risk of damage in these regions is significantand resulting damage is usually either minimal (dykes hold) or catastrophic(breakage, with total submerging; e.g. California 1997; China 1998).

A striking irony is that floodplains often are among the most desirable of farmlands. Many of these areas have vast hectares of aquic soils that typically requiredrainage for productive farm use. Extensive soil drainage systems have been built inriver and coastal plain flood regions, with notable examples in The Netherlands,south-central Asia, China, the Atlantic coast of the USA and floodplains of majorrivers, including the Mississippi in the USA and the Red River of Canada and theUSA. However, nearly all of these regions are vulnerable to floods, in spite of theextensive use of dykes, dams, land contouring and shaping, together with surfaceand sub-surface drainage systems. The economic cost associated with flooding inthese areas continues to escalate due to rising land prices, rising costs of maintainingimpounding structures, population pressures, higher costs of production and theintroduction of higher value agriculture. It is therefore important that future floodplanning looks not only to structural solutions but also to land-use planning, zoningand other solutions that encourage agricultural production in less vulnerable areas.

In this brief section attention is restricted to damage from heavy rainfall that occursin the absence of significant or widespread flooding. Soil erosion, disruptions tocritical agricultural activities, the lodging of crops, increased moisture leading toincreased problems with diseases and insects, soil moisture saturation and runoff,soil temperature reduction, grain and fruit spoilage and transportation interruptionsare among the more significant agricultural impacts from heavy rainfall.

As discussed earlier in the definition section, rainfall can be described as heavyfor several reasons. Convective storm systems typically produce short duration, highintensity rainfalls that can have significant erosive power on soil and damagestanding crops. Some convective systems are longer lived, however, and can beparticularly destructive. Tropical storms or hurricanes also produce very intenserainfall. Slow moving systems can dump enormous amounts of water over a periodfrom one to several days. These tropical systems are also accompanied by high winds,especially in regions where they first make landfall or on isolated islands, whichcompounds the problems created by heavy rainfall. Trees, crops and even buildingsare structurally weakened by saturated or flooded ground and are subject to blow-down from accompanying winds.

As mentioned, heavy rain on frozen or snow-covered soils is a significantproblem. In some cases, snowpack is able to absorb a vast percentage of theincoming rain, but if rains are significant and persist, the pack will begin to meltand will lead to substantial run-off problems.

Many agricultural systems are sensitive to any rainfall during critical periods,especially the maturation and harvest phases of certain crops. For instance, harvestsof grapes for raisin production and other fruit crops, depend on dry, sunny weatherduring the final harvest and drying period and this type of agriculture has developedin areas that have an extremely low probability of rainfall at this time of year. Anyrain during this period can be severely damaging. Heavy rainfall in these cases maybe less than 10 mm in a single day, or only 20 mm over several days. Similarly, anyrain after ball-bursting and during harvest of the cotton crop can be quitedetrimental to crop quality and yield.

5.2.2.2Mechanisms of heavy rain damage


Heavy rainfall that has a long duration (several days to several weeks) can delaycrop maturity and delivery to market. The weather systems that cause theseconditions usually cover a relatively large region. This may make for regionaldifferences in product delivery and reduce or eliminate normal market advantagesfor the affected region. In smaller, agriculturally-dependent countries, these weathersystems may cause delays and production reductions nationwide, creating significanteconomic hardship throughout the economy.

Usually, agricultural systems are well adapted to recurring (anticipated) heavyrainfalls appropriate for the specific climate of a region. For instance, throughoutmost of the interior north-western USA extreme 24-hour recorded rainfalls are lessthan 50 mm in most locations, even in a 100-year record. Conversely, many interiorsouth-western USA locations (many of which receive less annual rainfall than theirnorth-western counterparts) have recorded 24-hour rainfalls exceeding 120 mm, dueto a different climate regime (highly convective, tropical-origin air during thesummer months). Obviously, the design of agricultural systems, structures and otherinfrastructure is different in these two regions because of this difference inprecipitation.

There are important local to regional-scale factors that should be considered whenplanning for heavy rainfall and/or flooding. These include geographic factors (latitude, spatial dimensions of area of concern, culture and politics, economics,etc.) and topographic factors (elevation, slope, aspect, spatial scales, etc.). Thesefactors help frame not only how much precipitation will fall, for how long, and howfrequently, but also assist in determining its impact especially related to flooding.

For risk analyses and planning purposes hydroclimatological data typically areemployed. These studies are usually conducted using point data, however, there islimited availability of spatial climate data sets, particularly ones that are relevanton the timescales of heavy rainfall or flooding events. Thus, studies that examinethe topographic relationships with rainfall and flooding are needed.

For instance, maps of depth-duration-frequencies (ddf) have traditionally beenbased on point data, usually measured at relatively low elevation climate stations.Such maps may give an incomplete picture, however, of true ddfs in mountainousregions. Consider the data collected from two sites, only 20 km apart, on a 2 502-kmwatershed in Idaho, USA, as part of a study of precipitation characteristics in high

relief areas (Hanson and Johnson, 1997).

The low elevation site receives 275 mm of annual precipitation, 22 per cent inthe form of snow, while the high elevation site receives more than 1 100 mm ofprecipitation annually and 76 per cent in the form of snow. The NOAA Atlas

5.2.3GEOGRAPHIC AND TOPOGRAHIC

CONSIDERATIONS


Table 5.42- and 100-year return period

precipitation depths (mm) for givendurations

2-year storms

Duration Low elevation site (1 200 m) High elevation site (2 170 m)

(hours) Observed NOAA Atlas Observed NOAA Atlas

0.5 9.7 9.0 9.6 9.01.0 12 12 12 116.0 18 16 31 1824.0 26 23 69 32

100-year storms

Duration Low elevation site (1 200 m) High elevation site (2 170 m)

(hours) Observed NOAA Atlas Observed NOAA Atlas

0.5 31 24 29 251.0 35 31 29 326.0 47 46 53 4824.0 60 66 140 71

values refer to the US Department of Commerce’s NOAA Atlas 2 publication,which contains standard reference maps of ddf for the western US (Miller, et al.,1973). Clearly, the NOAA Atlas values are based on standard, mostly low elevation(dry), climatological stations that fail to capture the significant elevational changesin precipitation. from Table 5.4, it is noteworthy that for durations up to about sixhours the differences in the observed versus Atlas-derived values are minimal atboth 2- and 100-year return periods. However, for longer durations the differencesbecome large. For 24-hour duration storms, both the 2- and 100-year recordedprecipitation depths are approximately twice as great as the NOAA Atlas-derivedvalues at the high elevation site (they are nearly the same at the low elevation sitefor all durations and return periods). For agricultural and hydrologic planningpurposes these differences at longer durations are extremely important tounderstand. The NOAA Atlas 100-year storm depth is actually equal to theobserved 2-year storm depth at the high elevation site.

This is an important example of the type of analysis that is required for properagricultural management and placement. It points out the necessity of ascertainingtopographic and geographic influences on heavy rainfall and consequent flooding.The message is clear – be aware that information recorded at one location may notbe applicable for planning purposes at another site, in spite of their close proximity.Applicability lessens with increasing topographic and geographic complexity in a region.

Geographic Information Systems (GIS) can be extremely useful tools in theanalysis of flood-prone areas. A GIS in conjunction with digital elevation model(DEM) data can quickly determine slope and aspect of a region and can be used toprovide geospatial analyses of multiple spatial layers (elevation, slope and aspect –all at various scales or resolutions – along with soil characteristics, precipitation,temperature, vegetation, and other factors). GIS are being used to develop newfloodplain maps (at various frequency and severity levels) and delineate wetlands inmany regions and countries, using these types of spatial layers as well as others,including aerial photos. Such information will certainly assist in the best design ofagricultural systems, while accounting for reasonable risk.

A precursor to any substantive and useful planning for flood and heavy rain mitigation is an adequate database of meteorological, hydrological, agricultural,economic and other relevant information. Without thorough documentation ofpast events it is often difficult, if not impossible, to anticipate future conditions.

Reliable instrumentation for measuring rainfall, other types and forms ofprecipitation (including snow water equivalent), streamflow, lake and reservoirlevels and soil moisture and temperature is absolutely essential for monitoring andunderstanding the impacts of heavy rain and flooding. For heavy rainfalldocumentation, hourly (good), 15-minute (better), or breakpoint (best)observations are needed. In drier climates with less frequent heavy rainfalls, a 30- or 50-year record may be required before a sufficiently robust estimate of returnperiods can be determined. In wetter climates and/or those with relatively frequentheavy rainfalls, some idea of recurrence intervals and other statistics may beobtainable from just 10–15 years of record. Ground-based instrumentation shouldbe of sufficient quality to obtain accurate observations, even in very intense rainfalls.

Traditional, ground-based observation networks should be spatially denseenough to capture the horizontal dimensions of extreme storms. In regions whereconvective storms with relatively small horizontal size predominate, rain-gaugesshould be placed at key locations (often in and near important watersheds andbasins) at spacings that capture most storms.

In regions where available, the use of radar and satellite-derived fields ofestimated precipitation can be extremely useful. These images give a truer picture ofthe spatial complexity and variability of storm precipitation than available frommost conventional rain-gauge networks. They are not perfect, though, and ground-based data are usually needed to confirm and calibrate these remotely sensed rainfallestimates. These remotely sensed data fields are keys to timely and accurate heavyrain and flood forecasts and warnings.

5.2.4DATA AND ANALYSES FOR

ASSESSMENTS, PLANNING AND

MITIGATION


Statistical analyses can be performed on both point and spatial data sets. Anideal database management system would store temporally- and spatially-congruentdata from a variety of instrumentation and sources, including rain-gauge networks,radar and satellite imagery, standard weather networks, special agricultural andhydrological networks and others. It also would be ideal to have forecasts andwarnings stored. This ideal system would allow easy and timely analyses of extremeevents and would be able to generate both real-time and historical synopses in anydesired spatial or temporal framework.

Heavy rain idf statistics are essential for basic planning. Floodplain analyses(spatially, temporally) are also extremely useful. Both of these types of analysesshould be integrated into a usable GIS for ease of interpretation and presentationof results. Floodplain maps with appropriate information about probabilities (returnperiods) of certain amounts of precipitation and/or depth of flooding water should bedeveloped and used in risk assessments and agricultural planning. Based on theseanalyses, economic studies can be conducted to determine if certain types ofagriculture are appropriate and justified in certain flood regions. For instance, eventhough analyses indicate a region is subject to significant flooding every five years, itmay still be economically advantageous to farm the area, due to its high value ofreturn. Prudent planning would ensure that structures and other items that would bedamaged or destroyed by these frequent floods would not be built in the flooded area,but in less vulnerable areas nearby.

Hydroclimatological statistics based on recent history (typically 10–100 years)are thus extremely useful for planning purposes. Even very short records can provideimportant information, especially if the data are from critical and/or data scarceregions, and users are aware of the data and interpretation limitations. It must becautioned, however, that these statistics may sometimes give an incomplete ormisleading view of the future. In many cases, reliable hydrometeorological recordsare no more than 100 years long and often much shorter. Statistics about lowfrequency events (e.g. 100 or 500 year floods) can thus be in error due to therelatively short record upon which they were based. Of concern, too, is thatconditions for the most recent decade or two may not adequately reflect the comingdecade. Recent climatological studies hint at changes in storms and return periods,such as the work by Karl, et al. (1996), which found an increase in the proportion ofextreme (>50 mm/day) precipitation events during the time period 1900–1994 overlarge regions of the USA. In most cases, though, historical information and analysesderived from them provide the best available data upon which agricultural planningcan be based.

This section provides recent examples of the type and scope of agricultural damageinflicted by flooding and/or heavy rainfall. These examples were chosen based ondifferences in the causative factors leading to damage, in the type of agricultureimpacted and in the consequences of flooding and/or heavy rainfall – both short-term and long-term. The examples are:• Wintertime flooding due to snowmelt and heavy rainfall in California, USA,

January, 1997;• Late winter and spring flooding in the Midwest USA, 1997;• Summer (growing season) flooding in the upper Midwest USA, 1993; and• A sampling of flood events and impacts worldwide in 1995.

A good illustration of the advection of water downstream into (often lowland) agricultural regions is the fertile Central Valley region of California, USA. Meltingsnows and/or heavy rainfall in the Sierra Nevada mountains immediately to theeast sometimes transport significant volumes of water into the valley, in excess ofnormal spring runoff. This was the case in January 1997, when unseasonably warmand copious rainfall fell for several days, even at high elevations. All snowcovermelted in just a couple of days at elevations below 2 000 m and up to 800 mm ofrain fell in just 48 hours. The result was catastrophic damage to many agriculturalregions of the Central Valley. In several cases, dykes that had been built to with-hold river water in floodplain areas reclaimed for agriculture ruptured, putting

5.2.5.1Wintertime snowmelt/

rain-induced runoff and floodingimpacts on agriculture –California, USA (1997)

5.2.5EXAMPLES OF FLOOD AND HEAVY

RAINFALL IMPACTS ON

AGRICULTURE FOCUSING ON THE

USA. ASSESSMENTS, PREDICTION,

WARNING, MONITORING AND

MITIGATION


thousands of hectares under water for days. The following reports, compiled andissued from several sources including the California Department of Food andAgriculture (CDFA) and the California Department of Water Resources (CDWR),underscores the significance of this flood event to agriculture and chronologicallydescribes the events leading to the January 1997 floods. Record streamflows wereexperienced on many of the Central Valley rivers and streams and were much morewidespread than experienced in 1986.

October precipitation in the northern Sierra was about 75 per cent of average, buta storm at the end of the month brought unseasonably heavy rains to the centraland south coast and in the San Joaquin Valley.

The first half of November was quite dry but a major storm during the thirdweek saturated the watersheds; the month ended with above average rainfall (120per cent in the northern Sierra).

The first 12 days of December saw major storms, with relatively high snowlevels, which pushed many major reservoirs into flood control operations and causedlocal flooding on some Northern California streams. In the Sacramento Valley,overflow into the flood bypass system began during the second week of December.

• 21–23 December – a major storm brought snow to low elevations in the mountains.• Flood control storage in the Sacramento Valley was maintained by releasing excess

water and inflow as it occurred. However, in the San Joaquin Valley, inflow to manyreservoirs exceeded the more limited downstream channel capacity and encroachment into flood control storage gradually occurred in late December.

• A few days after Christmas, computer models indicated a very large, warm stormwas building in the Pacific Ocean. Media attention gradually began to take note ofthe predicted storm.

• The predicted storm seemed to delay for a couple of days and media stories beganto question whether it was really coming.

• On 29 and 30 December, heavy rains began at relatively high elevations, primarilyin the Sierra, melting the snowpack and releasing additional runoff.

• The most intensive precipitation occurred on 31 December and 1 January in thenorthern Sierra and on January 1 and 2 in the southern Sierra.

• Inflow to Oroville Reservoir set a new flood peak of record on 2 January, rising toabout 300 000 cfs. The 3-day volume was also a record, at 167 per cent of Oroville’sflood control storage. Similar records were broken on many other Sierra rivers,including the Sacramento River at Shasta Dam.

• 2 January – flows on the Feather River below the Yuba River exceeded designcapacity by 5–10 per cent. A levee broke near Olivehurst.

• Estimated peak inflow to the delta from the Sacramento River system exceeded600 000 cfs.

• January 4 – the west levee of Sutter Bypass west of Yuba City failed, inundating anestimated 37 000 acres. The town of Meridian was subsequently saved by constructing a ring dyke around the town.

• 3 January – the Tuolumne River at Modesto reached a new peak of record, swollenby water spilled from a full New Don Pedro Dam. A similar spill occurred on theSan Joaquin River below Friant; a series of levee breaks occurred downstream fromHighway 99. The San Joaquin River Flood Control System was overwhelmed andcaused increased potential for levee failures.

• 4–6 January – all but one of the reclamation districts on the lower San JoaquinRiver, from the Tuolomne River to Manteca (Mossdale Bridge), were inundated byfloodwater after levees broke from the excess flow.

• Mid-January – most Sacramento River region flood control reservoirs regainedflood control storage; San Joaquin River region reservoirs only regained about halftheir flood space because of more restricted downstream channel capacity.

• 21–27 January – a new series of Pacific storms brought a new round of flood flow butrunoff was not as large. The Sacramento Basin reservoirs regulated runoff to withinchannel capacity, including the two channels that had reduced capacity because oflevee breaks that were under repair.

Chronological hydrometeorologicalsummary


• 30 January – the lower San Joaquin River crested for the second time, setting a newrecord at Newman, but less than early January levels downstream at Vernalis andMossdale, where multiple levee breaks reduced the river stage.

• The Blue Canyon weather station (elevation 1 610 m) north-east of Sacramentoreported 1 900 mm of precipitation for December and January, (1 090 mm inDecember and 810 mm in January). December was the second wettest on record,only exceeded by December 1955 which saw 1 150 mm and was also a flood year.The December–January total was a new record. The two-month average at this station is around 625 mm.

After the initial flood emergency response by many local, state, and federal agencies the Department of Water Resources (DWR) directed short-term effortsfirstly to facilitate requests for the United States Army Corps of Engineers(USACE) assistance to repair damaged flood control facilities, and secondly tocontinue coordination of reservoir operations in the Central Valley to minimizefurther flood damage. Numerous levee breaks on both systems necessitated anextraordinary effort to develop and implement a short-term strategy for bringingthe flood control system up to at least a 20–25 year level of flood protection.

Local Reclamation Districts and the DWR quickly assessed the most criticalelements of the Sacramento-San Joaquin flood control systems and, whereappropriate, requested the USACE – under its authority – to provide flood fight andlevee rehabilitation. An unprecedented level of cooperation between DWR andUSACE in responding to local requests resulted in agreements made literally inhours and contracts being prepared and let sometimes on the same day.Consequently, several levee breaks were temporarily repaired and many werebrought back into service, albeit at a reduced level of protection. During the weeksfollowing the initial floods, continued high water levels in the rivers saturated leveescausing further damage – in some cases requiring emergency flood fights by theUSACE – and in most areas a heightened alert level was required to monitor thecondition of levees.

The January 1997 floods revealed deficiencies in the flood telemetry network.Flood forecasters were without critical stream and river flow information at severalkey locations during this event. These gaps in the existing flood network can befilled quickly to provide better warning and to better coordinate reservoir floodcontrol releases in conjunction with uncontrolled runoff entering the major riversystems. Many of these sites have existing stream gauges, but need telemetryequipment to send data to the California Data Exchange Center computers.

Due to flood control system limitations caused by levee failures, anextraordinary effort to coordinate reservoir operations throughout the CentralValley was required on a daily basis to safely carry flood control releases fromupstream reservoirs. This effort, carried out through the State-Federal JointOperations Center, and the USACE, in cooperation with San Joaquin ValleyReservoir operators, was successful in limiting further damage to areas downstream ofreservoirs. The high level of encroachment of most San Joaquin River systemreservoirs necessitated a continued high level of coordination among reservoiroperators for the remainder of the rain and snowmelt season.

The DWR was also directed by California Governor Wilson to assist localagencies in removing residual flood waters from inundated delta islands and othercatastrophically flooded areas within the Sacramento-San Joaquin flood controlsystems that remained after emergency levee repairs. This effort was essential toprotect public health and welfare and the integrity of the remaining flood controlsystem. The interior portions of the levee system were not designed to withstandwave wash and saturation from standing flood water which can seriously deterioratealready stressed levees and lead to further levee failures. Additional levee failurescould severely compromise the integrity of the flood control system and put morecommunities at risk of flooding, in addition to causing millions of dollars worth ofdamage to this key public safety infrastructure. The failure of delta levees wouldseriously jeopardize the range of options available to state and federal governmentsunder the Bay-Delta Accord and potentially disrupt the water supply of millions of

Short-term flood response actions


Californians in the Bay Area, Central Valley and Southern California. Residualfloodwaters continued to cause millions of dollars worth of agricultural damage, andaffected jobs and agribusiness-supported activities vital to many communities in theCentral Valley. Traditionally, the Federal Emergency Management Agency (FEMA)reimburses the cost of pumping delta islands as it recognizes the serious damagecaused by wave wash.

The February 1986 floods in the Sacramento Valley were a reference point forthe 1997 flood. For most northern Sierra streams, this was the flood-of-record priorto 1997. There was one major levee break, near the town of Linda on the YubaRiver, and five islands in the northern delta were flooded.

Immediately following the 1986 floods, the state asked the USACE to re-evaluate the Sacramento River levee system. This evolved into a five-phaserehabilitation programme. The first phase, upgrading levees in the Sacramento area,was essentially accomplished between 1989 and 1993. Recognizing the poorcondition of many delta levees, the California Governor and the Legislature workedtogether to develop a 10-year levee rehabilitation programme for delta levees. Morethan US$ 75 million had been spent by the late 1990s to rehabilitate and strengthendelta levees.

Flood damage to California agriculture from the winter rains in 1997 are summarized in Table 5.5, issued by the California Department of Food andAgriculture.

California Department of Food and Agriculture (CDFA) Secretary, Ann M.Veneman, testified before the Senate Budget and Fiscal Review Committee on29 January 1997 and released new figures estimating flood and rain damage to thestate’s agriculture at US$ 245 million. The figures were up from initial estimates ofUS$ 155 million, which were released on 10 January.

“These preliminary estimates are based on reports compiled by CountyAgricultural Commissioners”, said Veneman. Infrastructure damage, along with theimpact of flooding to land, private levees, farm equipment, buildings and irrigationsystems, appeared to be most significant at this time, with nearly US$ 124 million inreported damage. Damage to infrastructure, buildings, farm equipment, land,irrigation systems and private levees were significant long-term concerns.

According to the county reports submitted to the CDFA, 24 000 hectares ofcrops had been lost, with an additional 38 500 hectares damaged by the rain andflooding, at an estimated loss of US$ 89 million. Commodities most affected werewalnuts (US$ 16.8 million); livestock and dairy (US$ 16.5 million); nurseryproducts (US$ 16 million); alfalfa (US$ 15 million); wine grapes (US$ 13.8million); wheat (US$ 8.1 million); plums/prunes (US$ 6.1 million); and peaches(US$ 5.8 million).

Governor Wilson appointed a Flood Emergency Action Team (FEAT) to makerecommendations on the various aspects of the flood response efforts and futureflood protection needs in the state. In addition, the State Board of Food andAgriculture held a public forum on 6 February 1997 at CDFA Headquarters inSacramento to discuss the impact of the floods on California agriculture.

Various types of assistance were made available to farmers for disaster relief, and forthe development of mitigation strategies to deal with future floods in California.

The Federal Emergency Watershed Protection Program (EWPP) broughtassistance to many California farmers in 1997. Through the EWPP, the USDANatural Resources Conservation Service (NRCS) provided technical and financialassistance to prevent damage from flooding, runoff and erosion to safeguard peopleand protect property. Landowners who experienced severe property damage due toflooding were eligible for assistance. All projects required a governmental sponsor,such as a city, county or flood control district. Local sponsors of the EWPP wereresponsible for obtaining the necessary permits, providing 25 per cent cost-share andproviding for the operation and maintenance of completed emergency measures.

The Farm Service Agency (FSA) provided emergency funds for sharing withfarmers and ranchers the cost of rehabilitating farmland damaged by natural disasters

Government-sponsored reliefprogrammes for California

agriculturists, 1997

Flood damage estimates to Californiaagriculture in 1997


through the Emergency Conservation Program (ECP). Emergency practicesincluded debris removal, fence restoration, grading and shaping of farmland andrestoring structures. Cost-share levels up to 64 per cent were set by countycommittees. Eligibility for ECP was determined by county committees on anindividual basis. Cost-sharing over US$ 20 000 had to be approved by the DeputyAdministrator, Farm Programs. Technical assistance was provided by the NRCS.


Table 5.5Summary of January rain/flood

damage by commodity, April 1997

Through the Emergency Loan Assistance Program (ELAP), FSA providedemergency loans to help farmers cover production and physical losses in countiesdeclared as disaster areas by the President or designated by the Secretary ofAgriculture. The ELAP loans were made to farmers and ranchers who could notimmediately obtain commercial credit but who could provide collateral to securethe loan and were able to make the repayments. Borrowers were required to returnto conventional credit sources. Loan limits were 80 per cent of the calculated actualproduction loss and 100 per cent of the actual physical loss, or US$ 500 000,whichever was less.

Through the Non-insured Crop Disaster Assistance Program (NAP), FSAprovided crop loss protection to growers of commercial crops for which federal cropinsurance was not available. The NAP came into operation where the “area” hadsuffered a minimum of 35 per cent yield loss per crop. The NAP payments weremade to individual producers within these designated areas when individual croplosses were in excess of 50 per cent of the individual approved yield at 60 per cent ofthe crop’s average market price. Payments to individual producers could not exceed US$ 100 000 per year.

Loan guarantees available up to US$ 200 000 per business, or 95 per cent of theloan amount, whichever is less, were granted through the Loan Guarantee Program.Guarantees were provided to local lenders who made interim funds available tobusinesses at prime lending rates with no loan fees. Applicants must have appliedto the Small Business Administration (SBA) for disaster assistance in order to beeligible. Interim loan guarantees, or “bridge” loans were designed to bridge the gapbetween the application and receipt of funds from the SBA. Once the applicant’spermanent SBA loan was funded, the state guarantee was repaid. When the SBAfunded its loans quickly, the need for a bridge loan through the State of Californialessened.

The Farm Disaster Program (FDP) was designed for farmers whose crops wereimpacted by the 1997 floods and was a straight loan guarantee instrument with amaximum seven-year term at variable interest rates. Repayment of these loanscomes from normal operating income after farmers are able to return to profitability.The maximum guaranteed amount was US$ 500 000.

The following is a report outlining steps to deal with the significant impacts ofsevere flooding on agricultural land following the California floods in January of1997. It includes recommended strategies for dealing with the severe loss of soilnutrients during the flood and provides a perspective on the development of floodmitigation strategies in the future. It was written approximately one month following the end of the severe flooding.

FERTILIZING AFTER THE FLOODSDr A.E. Ludwick, Potash and Phosphate InstituteFebruary, 1997

Winter flooding is over, so it is back to business as usual... or maybe not. Recordrainfall in California will have an impact into the coming season and beyond. Theimmediate damage to trees, vines and winter vegetables is obvious, as is fielddamage caused by erosion. But there are also negative effects of wet soils and flooding that are not so visible.

How were soils impacted and what does this mean for future crops? Growers,crop consultants and fertilizer dealers need to understand how flooding affects theirfertilizer management decisions for 1997.

NitrogenDo not expect much nitrate (NO3) to be left in the soil profile. Whatever was

present in the fall most assuredly has been leached from the crops’ rooting zone ordenitrified (lost to the air as a gas). Either way, more nitrogen (N) fertilizer will berequired than normal to obtain expected yield levels.

Strategies for dealing with the impactsof winter flooding on California

farmland – the loss of soil nutrients


The ammonium form of N (NH4) is not denitrified, nor is it lost by leaching inall but very sandy ground. So some fall applied fertilizer N may have survived thefloods. The question is, had the applied ammonium (anhydrous ammonia, aquaammonia, urea, urea-ammonium nitrate solution, etc.) had sufficient time forsignificant quantities to be converted to nitrate before the soils became saturated?Time and temperature are important to this answer – more time and highertemperatures mean more of the ammonium has been converted to nitrate. Since soiltemperatures are warmer in October and November than in December, more N mayhave been lost by denitrification with the earlier application. If three or four weekselapsed from the time of fertilizer application to flooding with soil temperaturesmostly above 50°F, probably 50 per cent to all of it was lost.

Another source of N comes from crop residues and soil organic matter. Nitrogenis released as these materials decompose. The prolonged wet conditions will slowthe decomposition process, so this source may provide less than usual amounts of Nfor crop use.

Testing for residual nitrate in the profile is an especially good idea this year.Consult your soil testing laboratory for sampling details. Sample the whole rootingzone to obtain the complete picture.

PhosphorousPhosphorous (P) does not leach from soil as does nitrate, but is lost through

erosion of fertile topsoil. Also, reduced microbial activity and chemicaltransformations in saturated soils reduce P availability.

Most crops have a beneficial fungus called mycorrhizae colonizing their rootsystem. This fungus enhances phosphorous absorption by crop roots. Mycorrhizae isoften depressed after flooding, resulting in severe phosphorous deficiency infollowing crops. Mycorrhizae also influences plants’ abilities to take up zinc (Zn).

Prolonged flooding of soils causes several physical, chemical and biologicalchanges, some of which are not reversible. Phosphorous availability to plants isaffected by reactions with iron (Fe) and manganese (Mn), both of which are mademore reactive by waterlogging. As soils dry out, the forms of Fe and Mn phosphateschange, but the P availability remains low.

Slower organic matter decomposition, as mentioned for N, will also supply lessP than usual. And finally, eroded soils offer additional problems in that the organicmatter content will be lower, accentuating the potential for P deficiency.

Soil testing for available P is a generally reliable guide, at least with the Olsenbicarbonate test commonly used in California. However, P deficiency may be moresevere or more difficult to correct than in drier years. Higher rates of P fertilizer andadditional starter P are suggested to help overcome these conditions. Banding P tomaximize its concentration in the root zone could be especially effective this year.

PotassiumFlooding soils over the winter will not directly affect potassium (K) availability.

The exceptions are some K was undoubtedly lost through leaching of sandy soilsand some was lost from erosion of topsoil.

Reduced K availability will result when anxious growers return to their fieldsand attempt to work them while too wet, causing compaction. Compaction reducesavailability of K (and other nutrients) to plants. This is compounded by cool, wetconditions contributing to poor root development.

Besides the obvious nutritional benefit of supplying K in adequate amounts, Kalso enhances the crops’ ability to resist disease. It could be especially important tobuild up soil K for perennials weakened by prolonged flooding and especially proneto development of disease problems.

Soil testing will indicate a potential problem of K in soils that have beenflooded and eroded. Besides rebuilding soil test levels on leached soils, starter K isparticularly beneficial when soils are compacted, wet and/or cold.

There is no doubt that the 1997 season will offer many challenges for nutrientmanagement on soils damaged by flooding and erosion. It may be a cliché, but it istrue, “A fertile soil is not always a productive soil, but a productive soil is always afertile soil”. Many California fields are going into the 1997 cropping season with lessfertility than usual. Experience tells us that a programme of balanced fertilizer inputs


will likely give excellent returns. Don’t forget to consider secondary andmicronutrient needs as determined by soil testing. Nutrient management will needspecial consideration as we recover from the winter of 96/97.

A significant flood event occurred as a result of heavy rainfall and snowmelt in theMidwest USA in 1997. The event began with copious rainfall on saturated, frozen,semi-frozen or snow covered soils in early March and continued for more than twomonths. To provide a brief overview of the scope of the problem, including specific flood-related damage to mid-western agriculture, and to gain someperspective on mitigating strategies adopted or proposed, several brief news reportsduring and following the actual flooding are supplied here:

Flood Damage Assessed (Associated Press, 17 March). Officials are concernedflooding by the Ohio River could wipe out this year’s crop for some farmers becauseit stripped so much topsoil from their lands. The article outlines losses of equipment,infrastructure and livestock that resulted from the flooding.

President Seeks Flood Aid Funds (Reuters/Associated Press, 19 March).President Clinton asked Congress for US$ 2 billion to deal with the effects of anumber of natural disasters. The article outlines the funding sought, including US$126.1 million for USDA to deal with soil erosion and the problems the floodingcaused farmers and ranchers.

Potential Flood Impacts Examined (Bridge, 19 March). USDA weatheranalysts said they can’t tell at this time whether the predicted flooding in the Midwestwill affect crop planting. Although the flood threat to delay planting is very real, it isnot certain.

(Bridge, 19 March). The American Corn Growers Association said flooding inthe Midwest could result in corn prices quickly rising to US$ 4/bu.

(Des Moines Register, 19 March). While national forecasters Tuesday warned ofwidespread flooding across the upper Midwest this spring, a slow thaw had reducedthe risk in Iowa.

(New York Times, 20 March). Wheat prices surged on fears flooding in thenorthern Plains may delay planting.

Preliminary Damage Assessments for seven counties and two Indian reservations inwestern North Dakota are under way for flood damage. The assessments are being con-ducted in Dunn, Grant, Hettinger, Mercer, Morton, Sioux, and Stark counties as wellas the Fort Berthold Indian Reservation and Standing Rock Sioux Indian Reservations.

Several rivers in the Missouri basin have or will crest above flood stage causingdamage to numerous communities and their infrastructure. The Missouri River nearWilliston (Williams County) crested at 24 feet yesterday, four feet above flood stage.The National Weather Service (NWS) expects the river to remain at that level forsome time due to high runoff in eastern Montana that eventually feeds into theMissouri River.

Flood waters along the Cannonball, Heart and Knife rivers are beginning torecede as runoff flows towards the Missouri River. Yesterday, the Cannonball Riverat Regent (Hettinger County) was at 8.42 feet, well below its 22-foot flood stage.The Heart River at Mandan (Morton County) was recorded at 15.6 feet, 1.4 feetbelow the flood stage of 17 feet. The Knife River at Golden Valley (Mercer County)was measured at 15.29 feet, 8.71 feet below flood stage.

The Cannonball River destroyed an 80-foot steel truss bridge north-east ofHettinger (Adams County) on March 25. Traffic has been detoured four milesaround the bridge. The emergency manager for Dunn County reports 80 roadbeds inthe county damaged by floodwater. In Mercer County, a Beulah resident reportedthat 200 gallons of fuel oil spilled and mixed with water in his basement, flooded bythe Knife River.

Impact of the flooding includes damage to roads, bridges and flooded basements.Evacuations are described as minimal with a few isolated families forced from theirrural homes. No shelters are reported open at this time.

In Grant County, a farmer reported 50 pigs drowned and a rancher said he is unableto feed his cattle because they are stranded by high water on the Cannonball River.

Examples of reports of flood damagein 1997 including the western North

Dakota flooding, 27 March 1997

5.2.5.2Flood damage to the USA cornbelt, late winter to spring, 1997


The head of the Federal Emergency Management Agency (FEMA) announced todaythat federal disaster aid has been made available to help people in five southern Illinoiscounties recover from the effects of the recent Ohio River Valley floods.

FEMA Director, James Lee Witt, said President Clinton authorized theassistance under the major disaster declaration issued for the state this afternoonbecause of damage to private property from severe storms and flooding that struckthe state starting 1 March. Similar aid was ordered earlier this month for the flood-plagued states of Indiana, Kentucky, Ohio, Tennessee and West Virginia.

Immediately after the President’s action, Witt designated the countiesAlexander, Gallatin, Hardin, Massac and Pope eligible for federal funding tosupplement the recovery needs of stricken residents and business owners.

The aid, to be coordinated by FEMA, can include grants to help pay fortemporary housing, minor home repairs and other serious disaster-related expenses.Low-interest loans from the Small Business Administration also will be available tocover residential and business losses not fully compensated by insurance.

Under the declaration, Witt said federal funds also will be available on a cost-shared basis to the state and affected local governments in the five designatedcounties for approved projects that reduce future disaster risks. He indicated thatadditional forms of assistance and more counties may be designated later based onthe results of ongoing damage assessments.

This heavy rainfall and flood event was centred on the Upper Midwest USA statesof Iowa, Minnesota, Illinois and Missouri. It was different from the above-mentioned floods, with the heaviest rainfall and worst flooding occurring duringthe growing season. For this reason, the impact was more immediate and direct. Insome cases the flooding interrupted spring planting, while in other cases the heaviest rains arrived after plant establishment and interrupted growth through thecritical pollination phase, and into harvest in some locations. It was by far the mostsignificant agricultural rain and flood event, measured in economic as well as meteorological terms, during the 20th century in the USA.

This was a rain-driven flood. Rainfall totals for most of the region were thelargest of the 20th century for the 2-, 3-, 4- and 12-month periods that encompassedpeak summer rainfall months (Kunkel, et al., 1994). Estimated return periods formost of these totals were over 200 years.

Several factors were identified as chiefly responsible for the unprecedentedflooding. Rainfall totals for up to two months were more than 100 mm/week, whichprevious studies identified as a critical threshold for flooding. Abnormally cloudyweather reduced evaporation and kept incoming solar radiation and maximumtemperatures at unseasonably low levels. Preconditioning was significant, with mostof the hardest hit region reporting above normal soil moisture levels and highstreamflows prior to the event (early June). The spatial dimensions of the heavyrainfall were extreme, covering most of the four states. Nearly all of the rainfall thatwent into runoff eventually flowed into the Mississippi River, which reported itshighest river levels in history from Cairo, Illinois, to the Quad Cities of Illinois andIowa. The axes of heaviest rainfalls were often aligned along major riverways,including the Missouri and the Mississippi.

Meteorologically, the wet pattern was extremely persistent and abnormallystrong. For most of June and July, 1993 an unseasonably strong upper level troughwas centred just west and north of the impacted region. This resulted in copiousamounts of moisture being drawn northward from the Gulf of Mexico, along withabundant cloudiness and temperature reductions.

This flood and heavy rainfall event was catastrophic in proportion anddecimated much of the vast corn and soybean production area in the United States.In parts of Iowa, Minnesota and Missouri the persistently heavy rains andaccompanying floods were unprecedented since records began. Total agriculturalproduction losses in all affected states were estimated to be US$ 8.4 billion (thedirect losses), with total economic losses associated with agriculture (indirect as wellas direct) over US$ 18 billion (Changnon, 1995). The entire rural economy of these

5.2.5.3Flood and heavy rainfall damage tothe Upper Midwest, USA, summer

(growing season), 1993

Illinois flooding and disaster aid,21 March 1997


states was significantly affected. Crop insurance losses totalled more than US$ 1.7billion (Changnon, et al., 1997).

Insured losses are no trivial matter and are a growing concern where agriculturalinsurance is becoming more commonplace. Between 1991 and 1994, nine smallerinsurance firms in the USA became insolvent directly as a result of weatherextremes. Chief among these were the 1993 Midwest flood and the heavy rainfalland wind associated with Hurricane Andrew in southern Florida in 1992. Theinternational insurance companies have rapidly acquired an appreciation foragricultural extremes, especially flooding, and are becoming leaders in the analyses ofthese events and in planning for the future. In fact, these extremeagrometeorological events, chiefly driven by the 1993 flood, led to the creation ofthe Insurance Institute for Property Loss Reduction which initiated several keyactivities including a database of claims paid on catastrophes, and the developmentof databases relating weather perils to the potential for damage (Insurance Institutefor Property Loss Reduction, 1995; Changnon, et al., 1997). Clearly, well-documentedand thorough databases are essential for flood and heavy rainfall mitigation.

To conclude, an example of the widespread nature of flooding and heavy rainfalldevastation caused by floods is provided below. This is a synopsis of reports onflooding and/or heavy rainfall from around the world in just a six month period(late 1995), from Global Disasters. There can be little doubt that these extremeagrometeorological events are pervasive, and cause significant damage every year innearly every part of the cultivated world.

Azerbaijan5 October 1995Type: RainsStage: EvolvingAffected: Dead: 5; Homeless: 3 000Current activity: National response (early)Date of last report: 12 October 1995

Bangladesh24 August 1995 (03:47)Type: Monsoon rains/floodsStage: OngoingAffected: UnknownCurrent activity: International response (early)Date of last report: 24 August 1995

Benin10 October 1995 (12:12)Type: Torrential rains/floodsStage: EvolvingAffected: 86 000 peopleCurrent activity: International response (early)Date of last report: 10 October 1995

China10 August 1995 (04:30)Type: FloodsStage: EvolvingAffected: 11.1 million people; 2 million marooned, 3.04 million evacuatedCurrent activity: National response (advanced)Date of last report: 10 August 1995

Costa Rica10 October 1995Type: FloodsStage: EvolvingAffected: 3 262 personsCurrent activity: International response (early)Date of last report: 27 October 1995

5.2.5.4Flooding impacts throughout theworld – a short sample of events


El Salvador5 October 1995Type: FloodsStage: EvolvingAffected: 5 dead: 8 000 people incurred localized lossesCurrent activity: National response (early)Date of last report: 5 October 1995

Ghana25 September 1995Type: FloodsStage: AftermathAffected: No dataCurrent activity: International response (advanced)Date of last report: 25 September 1995

Guatemala9 August 1995 (10:41)Type: Floods/landslidesStage: EvolvingAffected: Approx. 7 100 peopleCurrent activity: National response (early)Date of last report: 9 August 1995

India9 September 1995 (12:13)Type: floodsStage: EvolvingAffected: More than 10 000 personsCurrent activity: National response (early)Date of last report: 9 September 1995

Korea15 July 1995 (10:27)Type: floodsStage: OngoingAffected: Missing: 70; Homeless: 100 000 families (500 000 persons)Current activity: Internationall response (early)Date of last report: 13 September 1995

Lao PDR19 September 1995 (06:25)Type: FloodsStage: EvolvingAffected: No informationCurrent activity: Nationall response (early)Date of last report: 25 September 1995

Morocco17 August 1995Type: Flash floodsStage: EvolvingAffected: Dead: 230; missing: 500Current activity: National response (early)Date of last report: 20 August 1995

Myanmar12 September 1995Type: FloodsStage: EvolvingAffected: 50 deaths and 15 persons missingCurrent activity: National response (early)Date of last report: 27 September 1995


Philippines7 September 1995Type: FloodsStage: AftermathAffected: 48 dead, 7 injured, 382 missingCurrent activity: National responseDate of last report: 11 September 1995

Somalia7 November 1995Type: FloodsStage: EvolvingAffected: 20 deadCurrent activity: National response (early)Date of last report: 7 November 1995

Sri Lanka7 October 1995Type: FloodsStage: EvolvingAffected: Approx. 20 000 peopleCurrent activity: National responseDate of last report: 11 October 1995

Thailand31 October 1995Type: FloodsStage: EvolvingAffected: 68/76 provinces (4.2 million people); 231 deadCurrent activity: National response (early)Date of last report: 8 November 1995

Turkey4 November 1995Type: FloodsStage: EvolvingAffected: 62 dead; 16 missing; 60 injuredCurrent activity: National response (early)Date of last report: 8 November 1995

Vietnam13 October 1995Type: FloodsStage: EvolvingAffected: 7 provinces; 108 dead; 316 homes floodedCurrent activity: National response (early)Date of last report: 13 October 1995

• AMS, 1970: Glossary of Meteorology. Huschke, R.E. (ed.), AmericanMeteorological Society, Boston, 638 pp.

• Changnon, S.A., Changnon, D., Fosse, E.R., Hoganson, D.C., Roth, R.J. andTotsch, J.M., 1997: Effect of recent weather extremes on the insurance industry: major implications for the atmospheric sciences. Bulletin Amer.Meteor. Soc., 78:425–435.

• Changnon, S.A. (ed.), 1996: The Great Flood of 1993. Westview Press, 319 pp.• Foth, H.D., 1978: Fundamentals of soil science. John Wiley and Sons, 436 pp.• Gommes, R. and Negre, T., 1992: The role of agrometeorology in the alleviation of

natural disasters. FAO Agrometeorology Working Paper No. 2. FAO, Rome, 22 pp.

• Hanson, C.L. and Johnson, G.L., 1997: Spatial and temporal characteristics ofprecipitation on the Reynolds Creek Experimental Watershed in southwestIdaho. In: Proceedings of the Workshop on Climate and Weather Research, USDepartment of Agriculture. Ag. Res. Service No. 1996-03, pp. 48–57.

REFERENCES


• Insurance Institute for Property Loss Reduction, 1995: 1994 Annual Report.Report No. 1, 45 pp.

• Karl, T.R., Knight, R.W., Easterling, D.R. and Quayle, R.G., 1996: Indices ofclimate change for the United States. Bulletin Amer. Meteor. Soc, 77:279–292.

• Kunkel, K.E., Changnon, S.A. and Angel, J.R., 1994: Climatic aspects of the1993 Upper Mississippi River Basin flood. Bulletin Amer. Meteor. Soc,75:811–822.

• Lowry, W.P., 1972: Weather and life. Academic Press. 305 pp.• Miller, J.F., Frederick, R.H. and Tracey, R.J., 1973: Precipitation-frequency atlas of

the western United State, Volume V, Idaho. US Department of Commerce,National Oceanic and Atmospheric Admininistratiuon, NOAA Atlas 2.

• Raper, C.D. Jr. and Kramer, P.J. (eds.), 1983: Crop reactions to water and temperature stresses in humid, temperate climates. Westview Press. 373 pp.

• Ritter, W.F. and Beer, C.E., 1969: Yield reduction by controlled flooding of corn.Transactions ASAE, 12:46–50.

• Shaw, R.H., 1977: Water use and requirements of maize. A review. In:Agrometeorology of the maize crop. Publication 481, WMO, Geneva, pp.119–134.

• Susman, P., O’Keefe, P. and Wisner, B., 1983: Global disasters, a radical inter-pretation. In: Interpretations of calamity, pp. 263–280. The Risks and HazardsSeries: 1. Hewitt, K. (ed.), Allen and Unwin Inc., Boston, 304 pp.

• WMO TD No. 836, 1997: Extreme agrometeorological events. WMO CAgMReport No. 73, WMO, Geneva, 182 pp.


No.

Que

stio

nA

rgen

tina

Arm

enia

Aus

tral

iaA

ustr

iaA

zerb

aija

nB

angl

ades

h

CC

Con

tine

nt c

ode

SAA

SA

UE

AS

AS

SCSu

bcon

tine

nt c

ode

SSA

SWA

AU

NE

SWA

SWA

MC

DM

ain

clim

ate

divi

sion

cod

eB

CC

CB

ASC

DSp

ecifi

c cl

imat

e di

visi

on c

ode

BS

CF

CF

CF

BS

AW

1D

oes a

gric

ultu

re a

nd/o

r liv

esto

ck in

you

r cou

ntry

get

affe

cted

by o

ne o

r mor

e of

the

follo

win

g ex

trem

e ev

ents

:1.

1D

roug

hty

yy

yy

y1.

2D

eser

tific

atio

ny

yy

ny

m1.

3Ex

trem

e ho

t and

dry

(he

at w

aves

, etc

.)y

yy

yy

y1.

4Ex

trem

e co

ld w

eath

ery

yy

yy

m1.

5Fr

ost

yy

yy

ym

1.6

Floo

ds (

incl

udin

g co

asta

l ero

sion

, sea

-lev

el ri

se a

nd in

unda

tion

,sa

liniz

atio

n, w

ater

logg

ing)

yy

yy

yy

1.7

Hea

vy ra

infa

ll (m

onso

ons,

etc.

)y

yy

yy

y1.

8St

orm

surg

esn

ny

ny

y1.

9H

igh

win

dsn

yy

yy

y1.

10Lo

cal s

ever

e st

orm

s (in

clud

ing

seve

re th

unde

rsto

rms,

hail

stor

ms,

snow

stor

ms,

ice

stor

ms,

torn

adoe

s, sa

nd st

orm

s, sq

ualls

)y

yy

ym

y1.

11Tr

opic

al st

orm

s (cy

clon

es, h

urri

cane

s, ty

phoo

ns)

yn

yn

my

1.12

Hea

t str

ess a

nd c

old

inju

ryy

yy

yy

m1.

13Fo

rest

and

bus

h fir

esy

yy

nm

m1.

14Lo

cust

and

gra

ssho

pper

inva

sion

sy

yy

nm

m1.

15V

olca

nic

erup

tion

s and

ear

thqu

akes

, inc

ludi

ng ts

unam

isy

yy

nm

m1.

16Ex

cess

ive

wat

er p

ollu

tion

ny

yn

mm

4H

as y

our c

ount

ry d

evel

oped

met

hods

/mod

els t

o:4.

1Pr

edic

t the

ext

rem

e ev

ents

?y

yy

nn

y4.

2M

onit

or th

ese

even

ts?

yn

yy

ny

4.2.

2A

re re

mot

e se

nsin

g te

chno

logi

es u

sed?

ym

yy

my

4.3

Prov

ide

war

ning

serv

ices

rega

rdin

g ex

trem

e ev

ents

?y

ym

yy

y5

Doe

s you

r cou

ntry

ado

pt a

ny m

etho

d(s)

to a

llevi

ate

or m

inim

ize

the

loss

to a

gric

ultu

re/p

astu

re/li

vest

ock

due

to e

xtre

me

even

ts?

yy

yn

my

6D

oes y

our c

ount

ry h

ave

a sy

stem

to q

uant

itat

ivel

y m

easu

re th

ein

tens

ity

of th

e ex

trem

e m

eteo

rolo

gica

l eve

nts?

nn

ym

ny

6.3

Are

thes

e in

stru

men

ts a

dequ

ate

to m

easu

re th

e ex

trem

e ev

ents

?m

mm

ym

y7

Doe

s you

r cou

ntry

hav

e a

syst

em to

ass

ess s

ocio

-eco

nom

ic im

pact

and

Ben

efit o

f the

ext

rem

e ev

ents

on

agri

cult

ure/

lives

tock

/fore

sts?

yn

mn

ny

8.4

Do

you

feel

it n

eces

sary

to c

ondu

ct p

ublic

aw

aren

ess t

rain

ing

onho

w to

cop

e w

ith

the

abov

e m

enti

oned

eve

nts i

n yo

ur c

ount

ry?

yy

ym

yy

9D

oes y

our c

ount

ry h

ave

a pr

oble

m o

f des

erti

ficat

ion?

yy

yn

yn

9.5

Any

aw

aren

ess t

rain

ing

cond

ucte

d ai

med

at c

omba

ting

des

erti

ficat

ion?

yn

yn

ym

CHAPTER 5 — THE IMPACT OF EXTREME WEATHER AND CLIMATE EVENTS ON AGRICULTURE – FLOODING AND HEAVY RAINFALL101A

ppen

dix

— T

able

A.1

Que

stio

nnai

re o

n ag

rom

eteo

rolo

gy r

elat

ed t

o ex

trem

e ev

ents

Yes/

No

resp

onse

s

Bar

bado

sB

eliz

eC

anad

aC

had

Chi

leC

olom

bia

Cze

ch R

epub

licD

enm

ark

Ecu

ador

Egy

ptFi

jiFi

nlan

dFr

ance

NA

NA

NA

AF

SASA

EE

SAA

FA

UE

E

CA

CA

NN

AN

AF

SSA

NSA

NE

NE

NSA

ME

AU

NE

SE

AA

DB

CA

CC

AB

AD

C

AW

AW

DF

BS

CS

AW

CF

CF

AW

BW

AF

DF

CF

yy

yy

yy

yy

yn

yy

y

nn

ny

yy

nn

yy

nn

n

my

yy

yy

nn

ny

nn

y

nn

yy

yn

nn

ny

ny

y

nn

yy

ny

yy

yy

ny

y

my

yy

yy

yn

yy

yn

y

yy

yn

yy

nn

nn

yn

n

yy

yn

nm

nn

nn

yn

n

yy

yy

ym

yn

ny

yn

y

yy

yy

yy

yy

ny

yy

y

yy

yn

ny

nn

nn

yn

n

my

yy

yy

yn

ny

ny

y

yy

yy

yy

yn

nn

yy

y

ny

yn

yy

nn

nn

nn

n

nn

yy

ny

nn

nn

nn

n

nn

yy

ny

nn

nn

nn

y

nn

yn

nn

yn

nn

yy

y

yy

yy

yy

yn

my

yy

y

yy

yn

yy

ym

nm

yy

y

yn

yy

yy

yn

ny

yy

y

yn

yy

yy

ny

ny

my

y

yn

yy

yy

yn

ny

yy

m

nm

ny

yy

ym

my

yy

m

mn

yy

yn

nn

nm

yn

y

yy

yy

yy

mm

yy

ym

m

nn

ny

yy

nn

yy

nn

n

nm

my

yn

mm

my

mm

m


Gab

aon

Ger

man

yH

unga

ryIn

dia

Iran

Irel

and

Isra

elIt

aly

Japa

nK

azak

hsta

nK

yrgy

zsta

nM

adag

asca

rM

alay

sia

AF

EE

AS

AS

EA

SE

AS

AS

AS

AF

AS

NA

FN

EN

ESW

AM

EN

EM

ESE

EASW

ASW

ASA

FSE

A

AC

CA

BC

CC

CB

BA

A

AW

CF

CF

AW

BS

CF

CS

CS

CF

BW

BS

AW

AF

ny

yy

yy

my

yy

yy

y

nn

ny

yn

my

ny

my

n

nn

yy

yn

my

yy

mn

n

ny

yy

mn

my

yy

mn

n

ny

yy

yy

my

yy

my

n

nn

yy

yy

my

yy

mm

y

nn

ny

yn

my

yy

my

y

nn

yy

mn

mn

yy

mn

n

nn

yy

my

my

yy

mn

n

ny

yy

my

my

yy

my

y

nn

ny

mn

mn

ym

my

n

nn

yy

yn

my

ym

my

n

ny

yn

yy

my

ym

my

y

nn

ny

mn

mn

nm

my

n

nn

ny

mn

mn

ym

mn

n

nn

yy

mn

my

nm

mn

n

ny

yy

ny

yy

yy

ny

n

nm

yy

ny

yy

yn

yy

y

nm

ny

nn

ny

ym

nn

m

ny

yy

yy

yy

yy

ny

y

nm

nm

nn

ym

yn

yy

y

nn

nm

nn

ym

yn

ny

y

mm

mm

mm

yy

ym

my

n

nn

ny

nn

nm

mn

nn

n

ym

ny

yy

yy

yy

my

n

nn

ny

yn

mm

ny

ny

n

nm

ny

yn

mm

mn

ny

m

CHAPTER 5 — THE IMPACT OF EXTREME WEATHER AND CLIMATE EVENTS ON AGRICULTURE – FLOODING AND HEAVY RAINFALL103

Mol

dova

Mon

golia

Mor

occo

Mya

nmar

Net

herla

nds

Nig

eria

Om

anPe

ruPh

ilipp

ines

Port

ugal

Rep

ublic

of K

orea

Rom

ania

EA

SA

FA

SE

AF

AS

SAA

SE

AS

E

EEEA

NA

FSE

AN

EN

AF

ME

NSA

SEA

SEEA

EE

CB

CA

CA

BB

AC

CC

BC

FW

CS

AM

CF

AW

BW

BW

AM

CS

CW

CF

yy

yy

ny

yy

yy

yy

yy

yn

ny

yy

ny

ny

my

yy

yy

yy

ny

ny

yy

ny

yn

my

nn

ny

yy

yy

yn

my

ny

yy

ym

yy

yy

my

yy

yy

yy

ny

ny

my

yy

yy

mm

ny

ny

mm

yn

ny

yy

yy

ny

mm

yy

yy

yy

yy

yy

my

yy

yy

mm

ny

ny

mm

yy

ym

my

yy

yy

mm

yy

yy

my

yy

ny

mm

yy

yy

my

yn

ny

ym

yn

nm

ym

ny

nn

mm

yy

ny

mm

yn

nn

my

yy

yy

yy

yn

mm

nn

yy

yy

ny

yy

mm

nn

yy

yy

ny

yn

mm

ny

yy

my

yy

yy

my

ny

yy

yy

yy

yy

yy

yy

yy

yy

yn

mm

yy

ny

yy

yy

nm

ym

mn

my

ny

yn

mn

nn

mm

nn

yy

ny

ym

yy

my

ym

yn

yy

yy

yn

ny

yy

yy

ny

ym

yn

mm

ny

yy

ny


Slov

akia

Solo

mon

Sri L

anka

Suda

nSw

azila

ndSw

itzer

land

Tha

iland

Turk

eyU

krai

neU

nite

dU

nite

dU

nkno

wn

Uru

guay

Isla

nds

Kin

gdom

Stat

es

EA

UA

SA

FA

FE

AS

AS

EE

NA

SA

EEA

USW

AN

AF

SAF

NE

SEA

ME

EEN

EN

NA

SSA

CA

AB

BC

AC

DC

CC

CF

AF

AW

BS

BS

CF

AW

CS

DF

CF

CF

CF

ny

yy

yy

yy

yy

yy

y

nn

ny

yn

ny

nn

nn

m

yy

ym

yn

yn

yy

yy

m

nn

ym

yy

yn

yy

yy

m

yn

ym

yy

ny

yy

yy

y

yy

yy

yn

yy

yy

yy

m

ny

ym

yn

yy

yy

yn

m

ny

ym

ny

nn

yy

yn

m

yy

ym

yy

yy

yy

yy

y

yn

ym

yy

yn

yy

yy

y

ny

ym

yn

yn

nn

yn

m

yn

ym

yy

nn

yy

yy

m

yn

nm

yy

ny

yy

yy

y

nn

yy

yn

yn

nn

yn

m

nn

nm

nn

yy

nn

yy

m

nn

ym

yn

yy

yn

yy

m

nn

ym

ny

yn

yy

yy

m

nn

yy

ny

yn

yy

yy

m

nn

yy

my

ym

yy

yy

m

yy

yy

yy

yn

yy

yy

y

yn

yy

my

yn

yn

yy

m

nn

yy

mn

yn

yy

yy

m

nn

ny

nm

ym

ny

ym

y

nn

nm

ny

nn

my

yy

m

my

yy

yy

yy

yn

ym

y

nn

ny

yy

my

nm

nn

y

mn

my

yy

mn

mm

nm

m

CHAPTER 5 — THE IMPACT OF EXTREME WEATHER AND CLIMATE EVENTS ON AGRICULTURE – FLOODING AND HEAVY RAINFALL105

No. of Yes’s Per cent of Yes’s No. of No’s Per cent of No’s No. of missing data Per cent of missing data

52 91.2% 4 7.0% 1 1.8%25 43.9% 28 49.1% 4 7.0%37 64.9% 14 24.6% 6 10.5%31 54.4% 19 33.3% 7 12.3%42 73.7% 10 17.5% 5 8.8%45 78.9% 5 8.8% 7 12.3%36 63.2% 16 28.1% 5 8.8%21 36.8% 26 45.6% 10 17.5%41 71.9% 9 15.8% 7 12.3%47 82.5% 4 7.0% 6 10.5%21 36.8% 24 42.1% 12 21.1%37 64.9% 10 17.5% 10 17.5%38 66.7% 10 17.5% 9 15.8%19 33.3% 28 49.1% 10 17.5%17 29.8% 29 50.9% 11 19.3%22 38.6% 24 42.1% 11 19.3%32 56.1% 21 36.8% 4 7.0%39 68.4% 13 22.8% 5 8.8%30 52.6% 14 24.6% 13 22.8%48 84.2% 7 12.3% 2 3.5%38 66.7% 12 21.1% 7 12.3%30 52.6% 19 33.3% 8 14.0%22 38.6% 12 21.1% 23 40.4%14 24.6% 32 56.1% 11 19.3%40 70.2% 5 8.8% 12 21.1%26 45.6% 27 47.4% 4 7.0%17 29.8% 15 26.3% 25 43.9%

Overall per cent of:Yes No Missing

56.3% 28.4% 15.3%

Total number of countries responding: 57


CHAPTER 6

HAIL, HIGH WINDS AND COLD INJURY(by T.I. Adamenko)

6.1 HAIL

Hail causes significant damage to agriculture – damaging crops, vineyards and fruittrees over large areas.

Hailstones are particles of dense ice falling from powerful cumulonimbus. Thephenomenon usually occurs during the warm season. The appearance of hail cloudsis associated with upward air flows, thermal convection or temperature differencesbetween the upward moving air and the surrounding air. Thermal convection, as aresult of the unstable stratification of the atmosphere, can be caused by the passageof a front or by the heating of an air mass by the underlying surface in intra-massprocesses. Hailstorms of the greatest intensity occur when both processes combine,for example, the passage of a cold front through a region with an unstable stratifiedatmosphere.

Hailstorms are usually accompanied by thunderstorms, showers and squalls. Hailoccurs with showers and thunderstorms in 40–45 per cent of cases and with squallsin only 7 per cent of cases. Hail is a rarer phenomenon than thunderstorms. Onaverage one case of hail is recorded for every 10 to 15 thunderstorms.

In the Ukraine 53 per cent of hailstorms occur with the passage of atmosphericfronts, the remaining 47 per cent of cases fall as a result of intra-mass processes. Insummer, hail usually occurs with the passage of fronts – 75 per cent with cold fronts,20 per cent with occluded fronts and 5 per cent with warm fronts. In April andSeptember the majority of hail is caused by intra-mass processes.

Hail is most detrimental to agriculture during the second half of the growingseason, the significant damage to winter cereals taking place during the maturationperiod. Spring cereals, however, are damaged at both grain maturation and at earlierstages of growth. Hail can ruin vegetable crops as well. Fruit orchards and vineyardssuffer from the adverse impact of hail during flowering, fruit formation and fruitmaturation.

The amount of damage depends on the size of the hailstones, their density,intensity of fallout and the type and stage of agricultural crop. Hailstones of 20 mmand more in a diameter always damage crops. Hailstones of 30 mm and more indiameter are able to destroy crops completely. The largest hailstones fall during thewarm part of the year.

The process of hailstone generation is dependent upon relief. Even on plains,damage done by hail is usually caused by air turbulence strengthening in theboundary layer above small hills and undulating land.

In Ukraine the most frequent hail fallout occurs in the mountainous regions ofCrimea and the Carpathians, where, on average there are 4–6 days per year withhail. Here it is vineyards that suffer most. On the plains in Ukraine the averagenumber of days with a hail is 1–3. South of the steppes experiences least hail. Heretemperature inversions formed under conditions of breeze circulation in the coastalzone make convection easier. Areas situated close to the coast and water reservoirssee a significant drop in the number of hailstorms recorded.

The frequency of days with hail over a territory is quite stable, varying littlefrom year to year.

In general, calculating the number of days with hail annually, by Poisson’s Law,accords quite well with actual data sets (see Table 6.1), though actual frequencyexceeds that given by Poisson’s Law in all cases.

Hail is observed mostly during the warm season. From April the number of dayswith hail gradually increases to a maximum in May or June on the plains. Inmountainous regions the greatest number of days with hail shifts to June or July.

During May–July the damage done by hail covers vast regions. After Augustthe number of days with hail reduces significantly. In winter hail is a very rare event.In mountainous areas hail occurs once every ten years during the winter months, inthe rest of the territory it take place even more rarely.

The duration of hail fallout can vary from a few to 15 minutes. The probabilityof such duration may reach 20–30 per cent in specific regions, and it amounts to 30per cent and more in mountain regions (see Figure 6.1). Long-duration hail fallout– more than 45 minutes – takes place very rarely. In southern regions long-durationhail does not occur at all.

Hail falls mainly in the afternoons when thermal convection development isat its maximum. The greatest probability of hail (25–40 per cent) is from 2–4pm(some regions 4–6pm) on the plains. In the foothills and mountains the frequency ofhail is greatest between noon–2pm (30 per cent of cases ). Hail is rare at night andin the mornings.


Calculation Number of days with hail (%)

0 1 2 3 4 5 6 7 8 9 10

KIEV (plains)

Actual data 22 29 22 11 8 3 3 1 1By Poisson 14 27 27 18 9 4 1

L’VOV (mountains)

Actual data 17 23 7 27 10 7 3 3 3By Poisson 5 15 23 22 17 10 5 2 1

Table 6.1Annual frequency of

days with hail

No.of April May June July August September Octoberhail days (%)

0 90 60 67 78 88 91 931 5 23 24 22 8 9 72 1 8 43 2 4 14 2 15 1

Table 6.2Kiev – frequency of days with hail

during the warm season

Figure 6.1Probability of hail at

various hours (%)

CHAPTER 6 — HAIL, HIGH WINDS AND COLD INJURY 109

During the warm season the probability of hail falling in one region or state is57 per cent; in two regions 30 per cent; and in five and more regions only 2 per cent.Intensive hail does not occur over the whole territory but in scattered showers. Hailis a local phenomenon. It mostly falls in isolated spots and occasionally in hailpathways. These pathways can extend over several hundred square kilometres. Thedistribution of hailstones over land is not uniform.

Hailstones of 20 mm and more in diameter and intensive hailstorms withsmaller sized stones cause the most significant damage economically and areextremely dangerous. In Ukraine small but intensive hailstorms occur mostcommonly (more than 70 per cent – see Table 6.3).

Usually, large hailstones fall from the end of April/early May to the end ofAugust/mid-September. In Crimea a fallout of large hailstones may occur at anypoint in the year, even winter.

An especially dangerous hailstorm occurred in western regions on 25 June 1969between 4–5pm. The fallout lasted 15–20 minutes. The diameter of hailstones insome cases was 100 mm. The hail destroyed and damaged agricultural crops overvast areas. The depth of hail cover on the ground reached 15 cm in some places. Onthe same day eastern regions of the country experienced hailstones of 80 mm indiameter falling for 10–15 minutes. This hailstorm was generated with the passage ofa cold polar front.

On the basis of data for Ukraine on hail frequency it is possible to define threetypes of hailstorm:

(a) Type 1. Hail covering small areas. This is the prevailing type;(b) Type 2. Hail exhibiting a scattered fallout pattern. The areas affected by hail vary

in shape and can range from ten to several hundred square kilometres. Such hail-storms may occur simultaneously in a number of regions; and

(c) Type 3. Hail falling in a strip. These strips vary in width from a few hundred metresup to ten kilometres. Sometimes strips can even extend a few hundred kilometres.

In summer hail occurs every other day in some region of Ukraine. In some yearshail falls every day in May. Hail rarely covers large areas.

In order to prevent the damage done by hail it is important to know thefrequency and range of hailstorms in a given territory. In the steppe zone of Ukrainethe number of days with hail is 7 within an area of 300 square kilometres. Whenthe area assessed is increased from 2 800 to 3 750 square kilometres, the number ofdays with hail increases from 18 to 24. When ever larger areas are included, there isno statistically significant increase in the number of storms.

Analyzing the density of the meteorological network is of interest. Such analysisreveals that 11 days with hail annually are recorded where there is onemeteorological station for 107 square kilometres; 24 days are recorded where there isstation for 16 square kilometres. With a further increase in the density of stations,the number of days with a hail becomes less. A network of one station per 7–10square kilometres makes it possible to record almost all occurrences of hail.

In the majority of cases of hail in Ukraine, westerly or northwesterly windsprevail (35–45 per cent of cases) and clouds are mainly cumulonimbus.

Hail occurs at a surface air temperature of 18–22°C. In mountain regions hailoccurs at temperature ranges of 10–14°C and 14–18°C (with 50 per centprobability). The probability of hail falling at low temperatures is less than 15 percent. At temperatures higher than 30°C hail very seldom occurs (8 per centprobability). Hail is accompanied by a significant decrease in temperature (6–10°C).

When spout clouds are generated, hailstones may be particularly large. This canbe explained by the strong turbulence within these clouds. The upward flows of airreach great heights and it is here where the hailstones are generated. These flowswithin a spout are whirling in nature. They are characterized by great power andstability, and are therefore able to keep hailstones at great heights for considerablelengths of time. The hailstones gradually build up new layers of ice. Thus, hailstonesfrom spout clouds are unusually large and fall in abnormally great amounts.

The hailstones of the greatest size occur in the central part of North America,where spout clouds are most widespread. Hailstones of 130–150 mm in diameter havebeen recorded. Such huge hailstones cause extreme damage to crops and orchards.

Hailstones characteristic Probability(diameter in mm) (%)

Small, intensive 71

20–29 15

30–39 10

>50 4

Table 6.3Probability (%) of

different size hailstones

Methods to counteract hailstorms centre around regulating the ice particleconcentration in clouds – in the zone of hailstone growth – attempting to restricttheir growth. In recent years much effort has been deployed in this direction and itcan be shown that in those areas which use anti-hail protection the damage done byhail is significantly reduced.

The most effective regulation of hailstone growth is carried out by the injectionof crystallizing reactants. To prevent the growth and fallout of hailstones it isnecessary to detect the hail centre in a cloud at the right time (using a radio-locator)and then to inject crystallizing reactants. Logistically, it is necessary to ensure thecapability of injecting reactants in any part of the region being protected. Theinjection of reactant into clouds is carried out in two ways: (i) bombarding clouds byartillery projectiles containing the crystallizing substance; (ii) using counter-hailrockets. Silver iodide and lead iodide are used as crystallizing reactants, whichpromote cloud growth rather than hailstone development. Effective anti-hailprotection is dependent upon reliable, accurate forecasting. Recent years havewitnessed a downward trend in the generation and fallout of hailstones of large size.

6.2 HIGH WINDSWind affects agriculture significantly. The degree of its effect on crops depends

on its speed, time of occurrence and duration.Wind with speeds of more than 15 m/s is classified as a severe weather

phenomenon; speeds of 25 m/s constitute an extremely dangerous phenomenon.Strong winds dry the upper layers of soil and plants. High wind speeds cause

increased transpiration in plants and can break the supply of water via the roots. Inthe case of high air temperature and low air humidity combined with high winds,plants dry up even if soil moisture is adequate.

High winds in spring hinder flying insects and bees decreasing possibilities forpollination. High winds with showers cause crops to be flooded, cutting of leaves,loss of flowers and fruits, and they hinder harvesting operations.

High winds in winter may encourage snow removal from fields if special barrierssuch as tree belts and windbreaks are absent. This can cause an irregular covering ofsnow. Snow is blown into ravines and gullies, the root systems of plants becomeexposed which can lead to frost damage. Re-distribution of snow is possible in highwind when the density of snow cover is less then 0.25–0.30 g/cm3.

High winds cause intensive evaporation from the soil which sharply reduces soilmoisture and lowers the water level in reservoirs. Water loss in this way can beeffected by relatively moderate winds.

Wind speed depends on the horizontal barometric gradient. It has beendiscovered that the mean barometric gradient for the day prior to the start of windstrengthening runs to 1.5 mb, and that at the moment of wind strengthening itexceeds 3.0 mb per 1 degree of a meridian. Wind speed is also affected by thephysical and geographical features of the territory: elevation, shelter features, etc.

Squalls are whirls with a horizontal component and short-term strengthening ofwind speed. They exhibit sudden changes of direction. Squalls are usuallyaccompanied by showers and thunderstorms, less often by hail. Local squalls whichcause crop destruction are classified as dangerous local winds.

Most parts of Ukraine have between 10 and 25 days of high wind per year. Inthe uplands to the east the mean annual number of days with high wind increases to40–50, and even 70–80 per year, in some places. South of the steppe regionexperiences 50–100 days.

The frequency of high winds varies from year to year. In 30–60 per cent of yearsthe number of days with a high wind deviates from the mean value by 1–10. Usinghistorical data it can be shown that the since the 1936 the number of days with highwind has been decreasing.

The mean duration of high winds varies from 2–12 hours. Those lasting lessthan 3 hours prevail in 60–80 per cent of cases. The greatest speeds in Ukraine aregenerally 20–30 m/s though in rare cases they can reach 40 m/s. High winds of thegreatest duration occur in the south and southeast of Ukraine. In Kherson, in


6.1.1MEASURES TO PROTECT

AGAINST HAILSTORMS

December 1946, the high winds lasted 143 hours, and the maximum speed of thewind reached 25 m/s. In the high mountainous regions of the Carpathians inFebruary 1965 the wind lasted more than five days, and its speed exceeded 40 m/s attimes.

The detrimental effects of wind are increased by other dangerous meteorologicalphenomena which occur when the wind is strengthening. In the warm seasonthunderstorms and showers often accompany squally strengthening of a wind. Insouthern and southeastern regions the high wind results in dust storms and hot drywinds. In the cold period the strengthening of a wind may be accompanied byblizzards and snow drifts.

High winds affect plants unfavourably. Most dangerous are winds with speeds above 10 m/s, bringing plenty of dust and destroying the surface soil layer. Thesewinds, called dust storms, are a phenomenon characterized by the transportation oflarge volumes of dust and sand by the high winds. During dust storms visibility isimpaired significantly. Their occurrence has both natural and human causes.Agricultural management, not ideally suited to the climatic zone, can encouragedust storms.

The drying of the upper soil horizon, absence of crop cover, low relative airhumidity (below 50 per cent) and absence of snow cover or ice crust in winter aswell as a poor cementation of soil and non-deep freezing constitute a complex rangeof factors which increases the probability of dust storms. This range of factors occursin the steppe zone. Dust storms mostly occur in spring, when the wind strengthens,and fields are in a ploughed state or contain poorly developed vegetation.

Soil structure, moisture conditions, presence of snow cover and relief areimportant factors in the formation of dust storms. When the soil moisture content inthe ploughed layer is 25 mm, dust storms may be formed at wind speeds above 15m/s. When this content is below 10 mm, dust storms occur even at wind speeds of8–10 m/s. In stable air conditions particles of sand and dust may be lifted up to greatheights under the influence of convective mixing.

Dust storms generally form between March and September, however, in thesouth and southeast they can develop in winter. The maximum number of days withdust storms usually occurs in the June–August period. A spring maximum is observedat some southern meteorological stations. This is caused by an early reduction insnow cover, an intensive rise in air temperature and the absence of good grass cover.

Dust storms may also happen in the steppe zone at the end of summer, whenthe soil dries up, and the fields begin to be ploughed after harvesting of the earlyspring cereals. Winter dust storms are a rare phenomena.

The upper soil layers in convex forms of relief begin to be blown away onwindward slopes with wind speeds of 8–10 m/s. The light soil particles move as adust great distances; the relatively heavier particles fall out and dislodge other soilparticles, which are then taken up by the wind. Thus, there is a “chain reaction”.

The intensity of the blowing of soil is proportional to the speed of the windraised to a third power. For example, when the wind speed changes from 12 to 15m/s, the intensity of the blowing of soil is increased approximately twofold

Barriers, such as tree belts and buildings, cause a slackening of the wind and theheavier particles to fall out, forming land-drifts. The lightest particles of soil canremain suspended in the atmosphere for a long time. And so, during a dust stormvisibility and light penetration worsen. Sunshine hardly gets through the dust-screen.

The horizontal extent of regions covered by dust storms varies markedly – fromseveral hundred metres to a thousand kilometres and more, and the vertical extentvaries from several metres to several kilometres.

Dust storms cause great damage to agriculture. They destroy and damage crops.High wind may remove significant volumes of earth, leading to reduced soil fertility.

In the steppe zone 1–5 days with dust storms are recorded annually. A higherfrequency (6–10 days) is observed in the southwest (Odessa, Crimea and Khersonregions), around Zaporozhye and Dnepropetrovsk to the east and in the Luganskregion in the southeast.


6.2.1DUST STORMS

In Ukraine strong dust storms were recorded in the spring seasons of thefollowing years: 1886, 1896, 1890, 1891, 1892, 1893, 1898, 1899, 1928, 1936, 1939,1947, 1948, 1949, 1953, 1957, 1960, 1969 and 1984. The most intensive were in1886, 1928, 1960. The wind erosion and deflation of soils caused by these duststorms were disastrous and resulted in huge losses to agriculture.

During the dust storm of 1886, in the east of Ukraine, the land-drifts reached2–3 metres and the soil depth decreased 25 cm in many places.

The dust storm of 1928 (26–28 April) caused great devastation over large areasof Ukraine. It affected the steppe and forest-steppe regions. The wind lifted morethan 15 million tons of dust to heights of 400–750 metres from an area of 1 millionkm2. Vast amounts of dust fell out over Ukraine, Romania and Poland. In areas,where the dust fell its depth ran to 12–25 cm. Incidences of dust deposition inDenmark, Sweden and Finland were recorded.

In the spring of 1960 wind speed reached 25–35 m/s. The low layers ofatmosphere were so saturated with dust that visibility decreased to 100–200 metres.The dry soil layer of 6–7 cm was affected by soil drifting. Transportation of the dustbeyond Ukraine’s borders was less than had been the case in 1928 despite the greaterwind speeds. This can be explained by the presence of forest strips. Poorly developedwinter cereals were killed in some regions and covered by small particles of soil inothers. More than 50 per cent of the winter cereal crop perished and extensive re-sowing was required.

In 1969 a dust storm of extreme force and intensity occurred in January andFebruary. It was a winter of exceptionally low snow levels. High winds demolishedthe upper dry layers of soil revealing tillering nodes. Crops poorly developed sincethe autumn, located in watershed sites and the slopes of non-forested fields weredestroyed; those on flat sites were buried. The tillering nodes were exposed 1–2 cmabove the surface. Exposed crops and those buried by dust drifts perished. Thephenomenon was accompanied with strong frosts. The loss of winter cereals reached62 per cent in some places.

The more structured the soil the more susceptible it is to blowing. Thereforewind erosion begins at different wind speeds for different soil textures.

Light-textured soils (sandy, sandy-loam, light-loamy) are most susceptible toblowing. These soils freely pass moisture to the deeper layers. Therefore, their surfacerapidly dries up and when exposed to relatively low wind speeds can be lifted up.The large particles of soil gather in dips in the relief or around obstacles such asbuildings, forest strips etc.

Table 6.4 gives the wind speed which causes wind erosion for different soil typesand textures.

The main agrometeorological factors which define dust storms occurrence arewind, soil moisture, presence of snow cover, relative air humidity and oscillation ofthe air temperature during 24-hour periods.

The likelihood of dust storms in the different seasons varies (see Table 6.5).Observational data shows that for eastern regions of Ukraine for the period1945–1996 dust storms in the winter and spring were continuous (more than 5 days)in 14–20 per cent of cases. These dust storms covered a large territory. In themajority of regions 80–100 per cent of the dust storms occurred under easterly andsoutheasterly winds.

During the summer and autumn dust storms tend to exhibit a local characterand occur less frequently under easterly and southeasterly winds. So, greater


Soil type Wind speed (m/s) at15 cm above a surface

Sandy 2–3Sandy-loam 3–4Light-loamy 4–6Heavy-loamy 5–7Clay-textured 7–9

Table 6.4Wind speeds causing winderosion for different types

of soil

attention should be paid to the most severe dust storms which occur in the winterand spring periods. In winter and early spring dust storms occur when there is littlesnow cover and low soil moisture. Blowing of soil in winter (black storms) takesplace when the relative air humidity is low and soil moisture is deficient. However,in some years dust storms in winter and spring take place even when there is goodhumidity in the upper soil layer, when there are strong daily oscillations in airtemperature and mechanical destruction and drying of soil can be observed.

Dust storms are often associated with droughts. Such a combination causesmajor losses to a national economy. Wind destruction or deflation of soils takesplace; crops are ruined as a result of seeds and poor developed seedling being blownaway; fields, roads, canals, water reservoirs, irrigation systems are covered by sandand dust; all kinds of infrastructure and communication networks are adverselyaffected. Wind erosion may remove soils huge distances. In 1960 dust from Ukraineand the Northern Caucasus of Russia was deposited in Romania, Bulgaria, Hungaryand Yugoslavia; and visibility was badly affected in Bielorussia, eastern Poland and atthe Baltic coast.

Dust storms – a negative effect of winds – are widespread all over the world.They occur in the central regions of western USA. In the 1930s they affected theGreat Plains from North Dakota to Texas. The dust storm of 1934 removedhundreds of millions tons of soil thousands of kilometres. On one day Chicago wasunder a dust layer of 12 million tons. Even remote New York experienced drought.Cereal yields decreased 75 per cent. Hundreds of thousands people affected by thedust storm left their homes and migrated west.

The largest dust storms occur in sandy deserts. Here they are responsible forputting large volumes of dust into the atmosphere and moving great masses of sandon the earth’s surface. As a result of selective blowing, soil with a high sand particlecontent is created and the organic element (the main source of nutrients for plants)decreases. Thus in places the dust storm causes loss of soil fertility and a change inthe chemical structure of the soil. The physical properties of the soil are changedtoo with changes in soil structure. In dry areas, steppes and deserts, dust storms areobserved usually in spring and early autumn in connection with dry soil.

Dust storms may be classified by the colour of the lifted dust – white, yellow,brown, red or black, depending on soil type.

The centre of a dust storm forms, at ground level, where the wind force issufficient to start wind erosion. Then it develops over a wider area. Dust storms canbe described as local or advective according to the source of their suspended sand-dust. Advective dust storms extend far from the centre of their origin. Satelliteimages show the movement of dust clouds from the Sahara over the Atlantic Oceanto Central America. Each summer 60–200 million tons of dust are deposited overthe Atlantic.

In the southern Sahara dust storms can cover areas of 2 500 x 600 km,stretching from the Senegal River to Lake Chad. They form latitudinal strips. Herewind speeds can reach 25 m/s and more.

Dust storms have exceptionally strong, sometimes disastrous, effects onKazakhstan and its neighbouring region. Wind speeds of 22–25 m/s and even 34–40m/s are recorded. These winds can be classified as hurricanes. In the aftermath ofsuch storms it is common to have complete fields with emerging seedlings coveredby sand.

Similar dust storms occur in the central states of North America. During longdroughts and high winds dust storms extend over enormous areas. Dust can be


Meteorological Dust storms (%)station

Under easterly and southeasterly winds 5 days or longer

Winter and spring Summer and autumn Winter and spring Summer and autumn

Svatovo 100 55 20 0Starobel’sk 91 70 14 14Belovodsk 93 46 14 0Lugansk 81 55 14 10

Table 6.5percentage of dust storms

1946–1996

transported to the Atlantic coast. In the USA the worst and most extensive winderosion occurred in 1933–1935. Crops were damaged in almost one-third ofcultivated lands in the country and millions of people were affected. In the mid-1970s 100 million acres of land suffered from wind erosion, 10 million acres of whichbecame practically unusable for agriculture. The situation was similar in Canada,where wind erosion of soils became widespread, especially in the central agriculturalbelt.

Winter dust storms do not reach the destructive levels of the autumn and springstorms. In the winter the snow cover protects the soil from blowing. In winters whenthe snow cover is light or absent and the soil is rather loose, dry and easily subjectedto deflation soil particles can be moved together with snow – snow-dust storms.Research shows that winter deflation is not particularly rare and there are regionswhere blowing of soil takes place all year.

To prevent or decrease losses to the national economy due to extreme naturalphenomena it is necessary firstly to detect the distribution and frequency of suchphenomena over a region.

In most countries, field afforestation is the main measure to protect the soil fromdust storms. Improving soil resistance to erosion can be achieved by careful selectionof cultivation methods, applying mineral and organic fertilizers, sowing grass andspraying various substances which enhance soil structure. It is also important toreduce the areas where dust can gather, especially in tracts in areas characterized byerosion. One major protection strategy is to establish well developed plant coverbefore the dust storms period. This can encourage a reduction in the wind speed inthe layer next to the ground by forming an effective buffer.

When assessing the impact of a dust storm on agricultural crops it is necessary totake into account the degree of development of the plants. On well-tilled crops thedeposition of soil moved by airflow is observed more often than soil carried by winderosion over long distances.

When looking at the conditions in which dust storms develop and data onstorm-induced damage it is evident that measures to reduce the wind speed at thesoil surface and to increase the hooking of soil particles are both crucial. Suchmeasures include the establishment of tree belts and windbreaks. Leaving stubblein fields, non-mouldboard ploughing, application of chemical substances promotingthe hooking of soil particles, soil-protective crop rotation using perennial grassesand seeding of annual crops are also important.

In planning and implementing protection measures it is necessary to take intoaccount the direction of prevailing winds, relief, microclimatic details and soilproperties.

In regions with intensive wind erosion, especially on wind-shock slopes or onlight soils, strip cultivation may be used. On fallow lands bare fallow strips of 50–100 m can be alternated with strips of grain crops or perennial grasses; springcrops can be alternated with winter crops. The direction of strips should beperpendicular to the damaging winds.

To reduce the oscillations of soil temperature and protect soil from wind erosionduring the winter–spring period cultivating without turning over furrows andretaining a cover of stubble from previous crops are two important measures.

6.3 EXTREME COLD WEATHER INCLUDING COLD INJURY

Temperatures of –10°C are extremely damaging to crops. In regions where wintercereals are grown low air and soil temperatures are the main causes of plant loss.

The low air temperatures experienced in Ukraine and in most of northernEurope are caused by the intrusion of polar air, usually via anticyclones from thenorth, northeast or northwest. Cold anticyclonic weather forms over Ukraine andpolar air is exposed to additional radiative cooling. Air temperatures may fall to–30–40°C.


6.2.2COUNTERACTING

DUST STORMS

Long periods of air temperatures below –10°C combined with othermeteorological phenomena are detrimental to many sectors of the economy,especially agriculture. Periods below –20°C can result in the loss of winter crops,fruit crops and vineyards due to frost injury. Frost injury is the most widespread typeof cold injury to affect crops.

Low soil temperatures at the depth of plant roots can cause frost injury. Suchreductions in soil temperature occur with strong frosts, in the absence of snow coverand with deep freezing of the soil. The soil temperature around the plant roots iscritical. Below a certain temperature irreversible processes take place in the planttissues, killing plant cells.

Most frost injury to winter crops takes place in the first half of winter beforesufficient snow cover which would afford protection has formed. In the second halfof winter frost injury happens in regions with unstable snow cover.

There have been many studies of plant injury caused by extreme cold weather.Under low temperatures basically a plant dries out. The protoplasm (the living partof cells) dies. This happens because the concentration of cell sap rises, the distancebetween macromolecules is altered and the processes of energy interchange aredisrupted. The toxic products of metabolism accumulate within the plant’s tissuesand it is this which causes the protoplasm to die.

Firstly, extremely low temperatures damage the leaf primordia. As a result offrosts they becomes brown and deformed, lose rigidity, and the shoot and sometimesthe whole plant perishes. Secondary younger sprouts suffer less from frost injury. Thedegree of damage depends on the intensity and duration of dangerous frosts as well asthe stage of the plant’s development. Damage to part of a plant does not alwaysresult in damage or destruction of the whole plant. A determining factor is thedegree of frost injury to a tillering node – if it is heavy the whole plant will prerish.

The temperature at which a plant perishes varies from species to species andwith the variety. For the same variety this temperature depends on a status of plantsin autumn as well as on changes to their frost-resistance under influence ofagrometeorological conditions during the winter period. Frost resistance in wintercrops is reduced if there are long, intensive thaws when the plant’s rest isinterrupted. Following a thaw, if the temperature falls abruptly crops perish atrelatively warmer temperatures than is the case if the decrease in temperature ismore gradual.

The depth of the root system and degree of tilling have a large effect. Well-tilled hardened plants with tillering nodes deeper than 3 cm withstand muchstronger frosts compared with underdeveloped plants in the early phases of theirdevelopment.

In the winter of 1968–1969, when the majority of winter crops in Europe werelost to cold injury many studies were undertaken. In 35 per cent of tests on plantsthe leaf promordia had been damaged or completely lost.

Tillering nodes of winter rye are the most hardy to frosts. Well developed andhardened crops in a condition of winter rest or dormancy can withstandtemperatures of –24°C and lower at the depth of the tillering node. The tilleringnodes of winter wheat are less cold-resistant. They perish completely at temperatureof –22°C. Plants of winter barley perish at temperature of –13–16°C. At thebeginning and end of winter plants perish at higher temperatures.

The winter crops most frequently destroyed by frost are those grown on uplands,where snow cover is less and the depth of soil freezing greater.

Very dry, dense and over-humidified soil conditions impact unfavourably on thecrops status and their dormancy. At optimum soil moisture the degree of thinningout of winter wheat as a result of injury by low soil temperatures is only 4.5 per cent.At insufficient soil moisture 26 per cent are lost and at excess soil moisture 48 percent perish.

The main agrometeorological factor influencing frost damage in winter crops islow soil temperature at the depth of the tillering node. Cooling to the criticaltemperature of frost injury, even for one day, and especially after a thaw, results inthinning out of crops. Long (three days or more) and intensive cooling causescomplete devastation.


The critical temperature at which frost injury occurs does not remain constant.It varies throughout the autumn–winter season decreasing through autumn to winterand then rising at the end of winter. Thus, frost-resistance of winter crops is low atthe end of winter and lost almost completely at the spring renewal of vegetation.Therefore, spring frosts of –10–12°C can injure plants.

To acquire high resistance to extremely low air temperatures winter crops needdry sunny weather with a well-expressed diurnal range in air temperature during theweek before entering the dormant phase. The air temperature range should startfrom +10°C in the daytime down to –1°C at night, followed by a graduallydecreasing daily average air temperature into negative values and transition to thewinter weather regime. If such weather is broken by significant warming, any gain inhardening is lost, and the injured plants go to winter. The critical temperature forfrost injury rises to –10–12°C and frost-resistance decreases.

From variations in air temperature and soil temperature around the roots it ispossible to predict the frost-resistance of winter crops, as well as the level of criticaltemperature attained by plants at the beginning of winter. For Ukrainian conditionsthe following relations between critical temperatures of frost injury and accumulatednegative soil temperatures at a depth of 3 cm for the period since crossing the meandaily air temperature from 0°C to –10°C at the beginning of winter can be given:

Accumulated negative soil temperature °Ñ 0 –25 –40 –60 –70Critical temperature °Ñ –13.5 –17.0 –17.5 –18.5 –19.0

These relationship between accumulated negative air temperatures and criticaltemperatures of winter wheat apply, if they are received in early winter and withsnowless conditions or snow cover not exceeding 2–3 cm.

In assessing the agrometeorological impact of low air temperatures on plants itis necessary to know not only the critical temperature, but also the length ofduration of the critical or low temperatures. If soil temperature at the depth of thetillering node is equal –20°C for 26 hours, then 43 per cent of the winter wheat cropwill be destroyed; after 36 hours 65 per cent will be lost and after 46 hours 96 per cent.

These data show the destructive impact of low soil temperatures. The absolutevalues can vary depending on the variety of the plant, degree of hardening, nature ofthe previous weather etc.

The temperature of the upper soil layer depends on air temperature and heightof snow cover above the frozen soil. Research has shown that the deeper the soilfreezing, the lower the temperature at the depth of the tillering node. Deep soilfreezing reduces the size of temperature fluctuations in the upper soil layers. Thisprotects winter plants from the harmful impact of sharp temperature fluctuations.

A drop in soil temperature below the critical level depends on many factors,the combination of which widely changes. The basic components are airtemperature, height of a snow cover and depth of soil freezing.

A key factor in protection from cold injury is stable air temperatures and snowcover throughout the winter. Thaws, resulting in packing or disappearing snowcover, worsen dormancy conditions and reduce or destroy the protective propertiesof snow cover. Long thaws can result in the renewal of vegetation, which isaccompanied by the consumption of carbohydrates and hence by an increase incritical temperature and decrease in winter-resistance.

In regions with unstable winters the frost-resistance of winter crops does notremain stable, however, after thaws frost injury seldom occurs.

Cold injury, including complete destruction, can take place at the beginning ofwinter when strong frosts occur before there is a good snow cover. At this time soiltemperatures drop below the critical level.

Strong frosts in mid-winter are often accompanied by winds which removesnow from fields, or they come after a thaw, when the snow has melted or stronglypacked. This can cause frost injury. Finally, frost injury can occur at the end ofwinter or early spring, when with dormancy over winter crops may be subjected toeffects of low temperatures.


The likely impact of frost injury can be interpreted from forecasts of extremelylow temperatures and the critical temperatures of frost injury for the various crops.This can be determined experimentally or calculated. Calculations are derived fromaverage minimum soil temperatures at a depth of 3 cm. The equation is as follows:

Tcr = 14.056 + 1.916t + 0.172t2

Tcr = critical temperature of frost injury (°C)t = average minimum soil temperature at the depth of the tillering node

The critical temperature marks the limit of a plant’s frost-resistance. Bycomparing this temperature with the actual minimum soil temperature at the depthof the tillering node, it is possible to predict the results of dormancy. If the criticaltemperature is lower than the soil temperature, frost injury will not occur. Lossesmay be significant, however, when soil temperature is equal to the criticaltemperature or lower. The ratio of the absolute minimum soil temperature at thedepth of the tillering node to the critical temperature is called the frost-risk factor.

K = t/Tcr

K = frost-risk factort = minium soil temperature at the depth of the tillering node (°C)Tcr = critical temperature (°C) resulting in injury to more than 50 per cent ofa crop

With meteorological data it is possible to determine a status of a crop at anygiven time.

The relationship between frost-risk factor (K) and injury to the crop (M) in percent is expressed thus: M = 77.94K79. Table 6.6 was compiled using this equation.

Conditions on the lands on the left bank of the Dnieper River during thewinter of 1968/69 were extremely unfavourable. At the end of January and intoFebruary air temperatures dropped to –29°C. Soil temperature at a depth of 3 cmwas –17–20°C; this was equal to or below the critical temperature for frost injury.Around 60–70 per cent of the crop was destroyed.

Unfavourable dormancy conditions result in retarded development and growthof injured plants in spring and summer. The initial formation of wheat ears is sloweddown. In spring the growth and development of injured plants can be seen, however,the subsequent differentiation of their generative organs may be abnormal, theprocess of ear formation may be damaged, flowers are underdeveloped and theoverall growth of the plant is slowed down. Such plants, even at optimumagrometeorological conditions in spring and summer, have much reduced yields.

Damage and destruction due to extreme cold weather in winter results in greateconomic losses – reduced grain harvests and great expenditure on re-sowing ofdestroyed winter crops.

The main measures to protect winter cereals from cold injury are retention ofsnow cover and introduction of frost-resistant varieties.

The average annual absolute minimum air temperature (Tmin) is a goodagroclimatic index of conditions for agricultural crop dormancy in Europe, especiallyfor climates with mild winters without a stable snow cover (most of Central andEastern Europe).

Snow cover significantly reduces the detrimental impact of low temperatureson winter crops. To make the correct predictions regarding conditions for wintercrops it is necessary to know the duration of standing snow cover. The number ofdays with snow cover varies regionally and is determined by the intensity of cyclonicprocesses. In the northern part of Western Europe the passage of cyclones at ratherhigh average temperatures in winter is fairly frequent. At Tmin –2°C the numberof days with snow cover is 20 days, at Tmin –15°C 40 days and at Tmin around–17°C it exceeds 50 days. In regions with a more continental climate such as easternHungary, Romania, Ukraine and the southern European part of Russia, the small


K M (%)

0.55–0.75 1–20

0.76–0.87 21–40

0.88–0.96 41–60

>0.97 <61

Table 6.6

number of days with snow cover is recorded at lower values of Tmin. Here 30 dayssnow cover are experienced at Tmin of –16–17°C and 40 days with Tmin of–18–19°C.

In the majority of cases strong frosts occur as a result intensive advection of acold air mass often accompanied by abundant snowfalls and followed by radiativecooling of air. Therefore in much of Western Europe the detrimental impact ofstrong frosts on winter crops is softened by the presence of snow cover during themost dangerous period.

For a correct assessment of the importance of low temperatures in winter it isnecessary to know the temperature at which crops are affected.

During the period of winter dormancy the frost-resistance of fruit crops isincreasing. They can withstand temperature down to –25°C and lower withoutdamage. Although it depends on the variety and the degree of hardening, thecritical temperatures which are detrimental to the majority of fruit with stones andpears are–25–30°C; many apple varieties can withstand temperatures down to–35°C. Some cherry species have great frost-resistance, though some peach andapricot species are very sensitive to low temperatures – temperatures between–20 –25°C being detrimental for them.

Depending on the variety and stage of growth, winter crops can withstandtemperatures down to –18–20°C without snow cover.

Minimum temperatures down to –10°C during the period of winter rest are, asa rule, not harmful to plants.

• 1971: Agroclimatic resources of the territory of socialist countries of Europe. Sofia,pp. 15–20.

• Budilina E., Prokh L., Snitkovsky A., 1976: Storms and squalls of moderate latitudes. Leningrad, Hydrometeoizdat, pp. 3–29.

• Volter B. Geis, 1997: No.2 12, UNESCO Courier, ISSN 0236-3879, pp. 11–13.• Golberg M., Volobuev G, Falei A, 1988: Dangerous weather phenomena and yeild.

Minsk, Uradjai, pp. 80–120.• Lebedev A., 1958: Climate of Europe. Leningrad, Hydrometizdat, pp. 10–90.• Logvinov K., Babichenko V., Kulakovskaya M., 1972: Dangerous weather

phenomena in Ukraine. Hydrometizdat, pp. 83–197.• Moiseichik B., 1975: Agroclimatic conditions and winter crops overwintering.

Hydrometizdat, pp. 5–278.• Nalivkin D., 1969: Hurricanes, storms and tornadoes. Leningrad, Nauka, pp.

10–38.• Lichikaki B., 1974: Winter crops overwintering. Moscow, Kolos, pp. 3–104.• 1991: Natural meteorological disasters in Ukraine and Moldova. Leningrad,

Hydrometizdat, pp. 17–138.• WMO,1971: Protection of plants against adverse weather. TN No. 118, WMO,

Geneva.• WMO, 1975: Drought and agriculture. TN No. 138, WMO, Geneva.• Robertson, G.W., 1983: Weather-based mathematical models for estimating

development and ripening of crops. TN No. 180, WMO, Geneva.• WMO, 1983: Guide to hydrological practices. Vol II. WMO, Geneva. • WMO, 1994: Climate variability, agriculture and forestry. TN No. 196, WMO,

Geneva.

REFERENCES


CHAPTER 7

LOCUSTS(by R.G. Gommes)

7.1 INTRODUCTION

WMO includes pests which devastate crops in the list of “agrometeorological disasters”, because their development is largely conditioned by agrometeorologicalconditions. The principal pest species in this group, for example locusts, grasshoppers, armyworms (Spodoptera, etc.) and the red-billed quelea (Queleaquelea), cause spectacular damage on a local scale, but are generally less damagingthan strictly climatic disasters.

There are very few global studies on the economic impact of these phenomena,and it is difficult to determine if the monitoring, or even large-scale controloperations, are economically justified (van Huis, 1993 and FAO, 1995). The relativeadvantages of prevention and control measures remain a debated subject.

A swarm of locusts consists of 40 to 80 million individuals, each of whichconsumes its own weight in plant material daily (approximately 2.5 g making a totalof 100 to 200 tonnes). If we consider that locusts don’t feed preferentially on crops(except off-season crops), that they develop more often during favourable conditionswhen both wild and crop biomass is flourishing, and that they feed mostly on foliage(rather than on grains), it is clear that even large swarms can only cause limiteddamage on a regional level. Local damage, however, can be serious.

7.2 DEVELOPMENT OF LOCUSTS

Various locust species are spread among several of the world’s semi-arid zones, inAustralia, Brazil, China and Africa, amongst others.

As they migrate in large swarms, their development is directly influenced bylocal and distant meteorological conditions. They go through several differentphases, each of which poses a threat to crops.

7.3 LOCUST-CLIMATE INTERACTIONS

Very simply: we begin with a situation where the locusts are found at low population densities. Under favourable conditions (low predation and abundantfood), populations develop until they reach a critical density at which individualsundergo morphological changes, often so dramatic that it is possible to mistakethem for a separate species. This is illustrated in Figure 7.1 for the case of the desertlocust found in the Sahel. Once populations have reached a sufficient size, theybegin to move under the influence of winds, up to 850 hPa1, until they reach anarea, often thousands of kilometres away, where conditions are favourable for reproduction (e.g. moist soils). This whole process takes place may times and thereare many generations within the area of the species’ distribution; statistically it finishes with a return to the point of departure.

This migration is largely controlled by agrometeorological, climatatological andsynoptic conditions:

(a) The geographic distribution is limited by climatic conditions and in general byenvironmental conditions (light, temperature, relative humidity and soil moisture

________

1 The utilization and monitoring of data at 850 hPa are quite exceptional in agrometeorology,

which normally confines itself to surface data.

and chemistry). Moisture is, in general, favourable for the growth of a number ofcryptogams and consequently damaging to many higher plants and insects. Locustsare no exception to the rule given in Gillon (1996) that high humidity is harmfuloverall. However, an input of water is necessary for several vital stages: ovogenesis, oviposition, the first days of embryonic development and rupture of diapause.

(b) Egg laying (oviposition) must be done into moist soil.(c) Gregarization is initiated when the population reaches a critical density

(“gregarization threshold”), which presupposes favourable feeding conditions (ormore favourable ones for the locusts than for their enemies). The locust has a paradoxical relationship with drought. According to Gillon (1996), locusts proliferate in arid regions, but only those with sufficient rainfall, while in temperate climates, it is summer drought which favours locust development.

(d) Swarms literally move under the influence of synoptic conditions and on a synoptic scale, blown by the wind once the latter has reached the rather modestspeed of 15 to 20 km/h. They can move 5 to 130 km in a day and cross the Red Sea(300 km) in several days. Exceptionally, as in 1988, locusts were able to cover the5 000 km separating Mauritania and the Caribbean. Solitary adults move mostly atnight, while swarms (gregarious phase) move mostly during the day.

It is interesting to note that, while many pest locusts are confined to semi-aridzones, it is often favourable rainfall conditions that promote the development oflarge swarms.

7.4 SOME LOCUSTS

The desert locust range is limited to the desert and semi-arid zones of Africa, theNear East, and south-west Asia which receive less than 200 mm of rain per year;this zone covers 16 million square kilometres over about 30 countries. During“scourge” periods, the area affected can reach 29 million km2; that is to say 60 countries, close to a fifth of the Earth’s surface and 10 per cent of world population2.These invasions do not occur cyclically, but are “pseudo-cyclic”. For example, inAfrica, the principal periods of activity have taken place in 1926–1934,1940–1948, 1949–1963, 1967–1969 and 1986–1989.

For the Australian locust (Hunter, 1996), swarming depends on rainfall occurringin the arid interior of the continent, which allows three or four generations todevelop in a year, and so reach the densities necessary for gregarization. However,

7.4.2AUSTRALIAN PLAGUE LOCUST

7.4.1DESERT LOCUSTS


Figure 7.1Transformation of a desert locust

between solitary and gregarious phases[illustration reproduced from the

Sécheresse glossary, 1996]

________

2 Most of this information is given on the FAO internet web page (www.fao.org/news/global/locusts/locFAQ).

Solitary phase form Gregarious phase form

Solitary phase colour Gregarious phase colour

should rain not fall, the eggs are very resistant to drought, and the gain in totalnumbers can be transferred to the following year when the eggs are “reactivated”(Launois, 1996).

The case of the migratory locust is very interesting and apparently paradoxical(Zibin Liu, et al., 1996), given that they develop during dry years. There is, in fact,an excellent correlation over a remarkably long data series which covers 21 centuries (Figure 7.2).

This can be explained as follows:(a) The main areas of reproduction in China are concentrated on the banks of rivers

and lakes, zones little used by agriculture because of fluctuating water levels. Thesemoist areas increase during dry years, thereby enlarging the zone suitable for locustreproduction. One method of locust control involves managing these zones.

(b) During a dry year, temperatures are generally higher than the normal, locusts aremore active, and three generations can develop under good conditions (instead oftwo).

(c) During a dry year, vegetation is more scattered, which leads to relative concentrations and therefore higher densities, and hence to the threshold of gregarization being surpassed.

7.5 MONITORING OF ACRIDIAN SITUATIONS

The FAO and other organizations regularly monitor the acridian situation in Africaand Asia. This includes both monitoring conditions favourable to locust development (e.g. rainfall in zones with little precipitation) as well as the appearance of vegetation in potential gregarization zones.

This poses enormous logistical problems, because of the low density ofmeteorological stations in arid and semi-arid regions, and the difficulty in estimatingrainfall by satellite in the absence of surface data. Even the satellites with thehighest resolution are generally difficult to use for this application.

It is often asked if satellites can be used to directly observe swarms. Earthsatellites do not have sufficient resolution. However, military satellites have veryhigh resolution permiting the identification and localization of swarms. Unfortunately, their utilization is hampered by a number of practical difficulties:

(a) Images are not normally available for civilian use;(b) High costs; and(c) The quantity of data to process is enormous.

7.4.3MIGRATORY LOCUST

Figure 7.2Frequency of drought and

migratory locust periods in thenorth and east of China fromthe second century BC to the

19th century AD (datareproduced from a chart in

Zibin Liu, et al., 1996)

Cen

tury

0 20 40 60 80 100 120 140 160 180 200

Frequency

DroughtLocust

19181716151413121110987654321

–1–2

CHAPTER 7 — LOCUSTS 121

There are a number of studies which attempt to evaluate the monitoring andcontrol operations for various locust species from an economic stand point, but nostudy has unambiguously concluded the utility of such operations, from this pointof view.

In practice, this aspect is highly important, but we should take into account thefollowing facts:

(a) Pest locusts and other migratory insect pests appear at very irregular intervals;(b) The semi-desert zones where swarms reproduce are very difficult and costly to

survey;(c) Control operations are damaging to the environment; and(d) Losses due to these pest species are, in practice, very difficult to evaluate

quantitatively.

• FAO, 1995: The fight against the locust. FAO, Rome, 16 pp.• Gillon, Y., 1996: Drought: an evil necessary for locusts? Sécheresse

7(2):133–144.• Hunter, M., 1996: The association of plagues of the Australian plague locust,

Chortoicetes terminifera (Walker) (Orthoptera: Acrididae) with sequences ofrain in the arid interior of Australia. Sécheresse 7(2):87–90.

• Launois, M., 1996: Drought adaptations of a wingless acridid in the BrazilianNordeste. Sécheresse 7(2):99–104.

• van Huis, A. (ed.), 1994: Desert locust control with existing techniques: an evaluation of strategies. Proceedings of the seminar held in Wageningen, TheNetherlands, 6–11 December 1993. Wageningen Agricultural University,Wageningen, The Netherlands. 132 pp.

• Zibin Liu, Wang, Q.C., Wang, H.C., Zheming Zheng, 1996: Drought and locustplagues in China: the case of Locusta migratoria manilensis (Meyen). Sécheresse7(2):105–108.

REFERENCES


CHAPTER 8

SPECIFICATION FOR A DATABASE OF EXTREMEAGROMETEOROLOGICAL EVENTS(by R.G. Gommes)

8.1 INTRODUCTION

Strictly speaking, extreme agrometeorological events include the direct and indirect impacts of extreme weather events on agriculture, taken in the broadestsense to include crops, livestock husbandry, fisheries and forestry. Extreme factorswill eventually lead to extreme agrometeorological events and disasters only if theyinteract with a vulnerable agricultural system.

The discussion which follows attempts to cover, at least from a methodologicalpoint of view, the impact of other extreme events as well, including some man-madedisasters such as chemical accidents and the consequences of earthquakes andvolcanic eruptions.

This approach is justified because of:(a) The existence of many borderline cases where non-agrometeorological events lead

to situations very similar to those traditionally falling into the province of agrometeorology, or the interaction of an extreme agrometeorological event withanother geophysical or other extreme factor. Examples are floods resulting fromearthquakes or dam failures, or fires resulting from earthquakes or meteoriteimpacts. More examples are provided below.

(b) Extreme factors often result from a chain of consequences of which only one has anatmospheric component, but the proper assessment of the phenomena requires thatthey be analyzed as a whole.

(c) The need to take into account not only direct and indirect atmospheric effects on production, but also their impact on tools, infrastructure and even general marketing conditions (access to markets) and weather-dependent price fluctuations.

We also suggest that the wording extreme agrometeorological events issomewhat contradictory in that many extreme events become “agrometeorological”only insofar as they affect agriculture. For instance, a tropical cyclone may be anextreme geophysical factor, but it will become an extreme agrometeorological eventonly if it hits an agricultural area and quantitatively or qualitatively affectsagricultural production. The unusual combination of moisture and temperature thateventually triggered the 1845/46 Irish potato famine (Bourke and Lamb, 1993)would not have been an extreme agrometeorological event had white potatoes notbeen introduced to Europe, etc.

By definition, a disaster is the result of the interaction between an extremefactor and a vulnerable system (Susman, et al., 1983), a definition which should alsolead to adopting consistent definitions of the related concepts of risk andvulnerability (Gommes, 1998).

A database of extreme agrometeorological events should thus more properly becalled a database of agricultural disasters resulting from extreme geophysical factors.

The purpose of a database of extreme agrometeorological events is, of course,to identify patterns of impacts on agriculture with a view to improving impactassessments, including impact forecasting, mitigation, adaptation and emergencyoperations whenever feasible. The proposed database is thus to be seen essentially asan operational tool.

The following section, therefore, starts with an attempt to list extreme factorswhich can potentially interfere with agricultural production.

8.2 CATEGORIES COVERED

It is obvious that a database of extreme agrometeorological factors must include ahierarchial typology of such factors, accompanied by a precise and quantitative definition. The hierarchial structure also makes provision for poorly definedextreme events included within a relatively broad category. Only after a proper typology has been defined will it be possible to provide a more complete outline for the structure of a database.

The following provides a tentative list of factors, which should be taken intoaccount, classified according to the highest categories of a potential typology, basedon the “geophysical” source of the disasters. Other approaches are possible, forinstance by detrimental factor, e.g. “mudslide” regardless of the cause of the mudslide(heavy rain, snow-melt associated with volcanic eruption, dam failure...). It wouldalso be possible to categorize events by the type of impact (famine, productionloss...), but this would pose some very serious, and possibly insurmountabledifficulties because the impacts are often based on very subjective and insufficientlydocumented assessments, particularly with regard to the extreme factor which ledto the disaster.

We also stress the fact that weather factors are correlated, so that, for instance,high sunshine and high temperatures usually occur together and need not be dealtwith separately.

This includes the natural atmospheric phenomena which directly harm crops bytheir instantaneous intensity or through longer term exposure. This covers theextremes (intensity, duration) of almost all meteorological elements: rain (torrential rain, drought, hail); strong winds (tornadoes, storms, tropical cyclones);and temperature (frost, heatwaves, high night-time temperatures). There do notappear to be obvious direct effects of high moisture. Lightning can be mentioned asthe cause of human and cattle death.

The incidence of hail is usually limited, although some extreme cases wererecorded. One such was the 1888 event in Uttar Pradesh, which caused about 250deaths. The “1888 blizzard” caused about 400 deaths in north east USA. The stormof 12–15 March 1993 on the Eastern Seaboard of the USA and Canada killed 300and affected 3 000 000 people, causing damage of US$ 1.8 billion. The storm, withrecord temperatures, wind, rain and snow, was nicknamed “the storm of the century”and is generally taken to have been worse than the “1888 blizzard”.

The meteorological elements to be listed here are almost the same as the ones givenin the section above, although the mechanisms of their detrimental action are different – rain (waterlogging of soils, floods, landslides, erosion); wind (abrasion bysand particles, soil erosion); temperature (increased water demand and resulting water stress, effect on sex differentiation of certain plants3); and moisture(incidence of diseases, conditions conducive to fires). Lightning is one of the causesof crop and forest fires.

Tropical cyclones constitute a perfect example of a complex interaction offactors, including strong winds, heavy rains, ocean spray, etc. Ocean spray is the saltwater blown inland which may salinize agricultural land. Storm surges are even moreharmful in terms of their impact on crop production. For a general account of salteffects in soils and irrigation water refer to Ayers and Westcot, 1976.

Unfortunately, sufficient information is not usually available on cyclones andtheir destructive power and the impacted agricultural system. For instance, cycloneHugo (17–23 September 1989) is well documented for the deaths and damagecaused in the north-eastern Caribbean and the south-eastern USA (the insured losswas US$ 4.9 billion). Though much property was damaged, it is not commonly

8.2.2INDIRECT NATURAL ATMOSPHERIC

FACTORS AND COMPLEX

INTERACTIONS

8.2.1DIRECT NATURAL ATMOSPHERIC

FACTORS


________

3 For instance for oil palm temperature at the time of differentiation of the flower primordia has an

effect on the frequency of female flowers – the only ones which will eventually produce oil three

years later. Another classic example is temperature-induced male sterility in rice.

known that most of the banana crop was also destroyed in Dominica, leading tolong-term suffering.

Valuable methodological conclusions emerge from a comparison of the July andAugust 1993 floods in the Mississippi, Missouri and Kansas river valleys (USA), the1981 Sichuan floods and the January and February 1995 floods in the Rhine andMeuse valleys. In the USA, although 1.7 billion ha were flooded, most damage toagriculture occurred as a result of the heavy rains in the whole basin throughwaterlogging of the standing crop. The Sichuan floods affected “only” 500 000 ha,but threatened the livelihoods of about 10 million people. The Rhine floodsoccurred outside the cropping season; they led to 250 000 people and most of theircattle having to be evacuated out of fear that the dams, many of them dating back tothe Middle Ages, would break. In comparison, the July–August 1998 Yangtze floodsaffected 26 billion ha of land, because the Chinese Government decided to breakdams and flood agricultural areas to protect downstream towns.

Man-made pollution4 is a multi-faceted issue, with many direct and indirect effects.The atmosphere plays a part in the formation of secondary pollutants (which resultfrom the reaction of normal atmospheric gases with pollutants) as well as in the dis-persion and transport of pollutants and their transfer between compartments(industrial plants → atmosphere; industrial plants → water bodies; atmosphere →soil; atmosphere → water bodies, etc.). Note that agriculture also constitutes asource of pollutants, either phytochemicals, fertilizers (nitrate pollution of thewatertable), manure and greenhouse gases. For an overview of atmospheric pollu-tion effects focusing on European forests, see ECE, 1997; for an analysis of thethreats to developing country agriculture, refer to Marshall, et al., 1997.

Atmospheric conditions usually play an important role too in that they areresponsible for the “contact” between pollutants and humans, farm animals andplants5. For example, high moisture increases the impact of pollutants such as ozone6

and acids; and high temperatures favour the accelerated decomposition of organicliquids (manure), a process which consumes oxygen and which can result in severeanaerobic conditions in water or waterlogged soils.

Stomatal opening, and the eco-physiological factors which control it, conditionto what extent pollutants such as ozone may enter and make contact withphysiologically active plant tissue (Kersteins, et al., 1992). According to theirnature, pollutants can affect plants by altering the environment (pH of soil, leachingof nutrients and mineral nutrients in general) or by physiological and biochemicalmechanisms.

Oil spills (either from platform or land wells, including during the 1991 Gulf War)have received a lot of attention by the media. Many, such as the Torrey Canyon(1967), Sea Star (1972), Amoco Cadix (1978), Ixtoc-I (1979), and Exxon-Valdez(1989; see Davis, 1996), have remained in the memory of people. Their effects tendto be localized and more detrimental to the environment than to fisheries.

A special mention can be made of the 1991 Gulf War oil spills and smoke fromburning wells. This caused massive air pollution and contamination of agriculturalland and water supplies throughout the Tigris and Euphrates valleys. Black-raindamaged crops over a wide area including Iran, Pakistan, Bulgaria and Afghanistan.

8.2.3.2Oil spills and well fires

8.2.3MAN-MADE FACTORS

8.2.3.1Atmospheric pollution as a source

of soil and water pollution

CHAPTER 8 — SPECIFICATION FOR A DATABASE OF EXTREME AGROMETEOROLOGICAL EVENTS 125

________

4 Not all “pollution” is man-made. An example is provided by rivers flowing through ore-rich

deposits, resulting in downstream heavy metal pollution.

5 A classic example of the combined effect of high moisture and toxic chemicals is the 1951 smog

which killed about 3 000 people in London.

6 Tropospheric ozone is probably one of the most agriculturally harmful pollutants in terms of

production loss (ECE, 1997).

Although nuclear accidents are difficult to hide due to the ease of detection ofradionuclides, they tend to be reported by governments only under media pressureand, as a result, data sets are not comprehensive.

Following the 1979 (28 March) nuclear meltdown at Three Mile Island(Pennsylvania), the 1986 (26 April) nuclear reactor accident at Chernobyl(Ukraine) eventually affected most of the immediate surroundings, Scandinavia andWestern Europe (except Spain and Portugal) to varying degrees. This constituted aperfect example of the transport of pollutants over long distances (Wirth, et al.,1987), their removal from the atmosphere and subsequent absorption by vegetationand the whole food chain (Marples, 1996).

A severe case of aquatic chemical pollution of the River Rhine affected Germanyand France in November 1986. This was caused by mercury-based pesticides, fungicides and other chemicals, which had originated from Basel (Switzerland). Foralmost two days, none of the governments along the Rhine knew the true nature ofthe chemicals flowing down Europe’s largest waterway. It was estimated at the timethat a 350-km stretch of the upper Rhine was “practically dead” and that it wouldtake 10–30 years for life to be restored. Similar estimates and statistics would beuseful in a database of agrometeorological disasters, if only to improve long-termimpact assessment methodologies.

The agricultural consequences of volcanic eruptions can be categorized as “global”and “local”, the magnitude of the former far exceeding that of the latter. Both largely depend on the main types of ejecta emitted. Volcanic eruption classifications range from Hawaiian (quiet eruptions with fluid lava) to Pelean7

(very violent eruptions accompanied by nuées ardentes and avalanches of explosivelava); Stiegeler, 1976.

The nuées ardentes (literally “burning clouds”) are high pressure and hightemperature gas and ashflows moving at speeds of up to 100 km/h. They transportlarge amounts of debris and pose very serious threats. The most violent type oferuption – Pelean – leaves very little chance to escape, burns all living creatures andresults in widespread destruction.

During 1783, over 12 km3 of lava and 500 million tonnes of noxious gaseserupted during the Laki Fissure eruption in Iceland (McGuire, 1997).

Major volcanic eruptions blow large amounts of dust and gases into the lowerstratosphere (15–25 km) where winds may distribute them over the globe in amatter of weeks or months. It is particularly sulphuric acid (derived from a combination of sulphur dioxide and water) that plays a significant role in loweringthe Earth’s albedo, usually resulting in lower surface temperatures. The effect maylast for several years, as illustrated in Figure 8.1, which shows stratospheric aerosolconcentrations between 1979 and 1995. During this time the following events

8.2.4.2Global effects

8.2.4OTHER GEOPHYSICAL FACTORS

8.2.4.1Volcanic eruptions

8.2.3.4Industrial accidents

8.2.3.3Nuclear accidents


Figure 8.1Distribution of stratospheric

aerosol in association withvolcanic eruptions. Based on the

attenuation of the 785 nmradiation at Rattlesnake

(46.4N,119.6W); data fromLarson et al., 1996

________

7 Named after Montagne Pelée, Martinique, which erupted on 8 May and 30 August 1902.

Esti

mat

ed st

rato

sphe

ric

aero

sol o

ptic

al d

epth

occurred: 1981 – Mount St. Helens eruption (USA), Alaid eruption (Aleutians);1982 – Nyamuragira (Zaire), El Chicon (Mexico); 1983 – El Chicon (Mexico);1986 and 1987 – Nevado del Ruiz (Colombia); 1988 – forest fires; 1991 – eruptionof Pinatubo (Philippines); 1994 – forest fires (south-east Asia). In the case ofPinatubo (June 1991), a well studied event because of its magnitude (McCormick,et al., 1995, Hansen et al., 1996), the global temperature decrease was about 0.5°C.

One of the best known historical examples is the “year without a summer”(1816 in the northern hemisphere) that followed the eruption of Tambora (Java)in 1815 (Stommel and Stommel, 1979). The eruption was widely used as a small-scale analogue of a nuclear winter (Sagan and Turco, 1990). Additional details canbe found in Schönwiese (1988), Briffa, et al., 1998 and de Silva, et al., 1998.

The local effects of volcanic eruptions can also be devastating, for example, no terrestrial species survived the eruption of Krakatoa on 26 and 27 August 1883. Theeruption caused more than 35 000 human victims as a result of the tsunami ratherthan the volcanic eruption directly (McGuire, 1997).

Effects tend to be relatively local only if the ash is not injected into the upperatmosphere. By way of an example, the 1989 (18 May) Mount St. Helens eruptionin Washington State reached Idaho and Montana where large quantities of volcanicash littered the soil to a depth of 1 m in places.

As shown by the eruption of Etna (Chester, et al., 1985) lava flows have thepotential to cause structural damage and will destroy any buildings in their path.But most important is the fact that prime agricultural land is rapidly “inundated”and becomes useless for agriculture and other related activities for hundreds of years.

A special mention should be made of the 1986 (21–24 August) “eruption” ofLake Nyos (north-west Cameroon) which was characterized by major CO2 and H2Semissions (Youxue Zhang, 1996). The toxic mixture caused about 3 000 deaths andin Nyos village only 2 of a population of 700 survived. Poultry and cattleexperienced heavy losses.

The already mentioned Pinatubo eruption on Luzon in the Philippines (1991)covered villages and agricultural land with sterile ashes, to the extent that around150 000 people were made homeless and 600 000 lost their livelihoods. The ashblanket reached a depth of several metres in the valleys close to the mountain,reducing to an average depth of 5 cm at a radial distance of 40 km. An estimated5 000 km2 was affected. Some of the most fertile land in the Philippines had to beabandoned, leading to immediate damage and loss of future income. Mudflowscreated havoc in flat areas up to 50 km from the crater, by, for instance, cloggingfishponds. An estimated 326 000 ha of forest, 43 000 ha of cropland and 16 000 haof ponds were damaged. Even in 1992, mudflows still occurred and buried crops(Rantucci, 1994).

On the positive side, it should be mentioned that where ash does not exceed 10 cm, it can be ploughed in and will increase productivity due to its pH, P, K, CAand Mg, even if Fe and S are excessive (according to Rantucci). Similarly, Besoain,et al., (1992), found the deposits from the Lonquimay volcano in Chile between1988 and 1990 had improved local soil quality.

Rantucci provides a detailed breakdown of the total loss incurred to theeconomy due to the Mount Pinatubo eruption (Table 8.1). Agriculture accounts for59.7 per cent of the total economic loss, most of it in the forestry sector and in theform of lost revenue.

Apart from disrupting infrastructure and destroying houses, earthquakes have littledirect effects on crops. However, they often lead to secondary extreme events causing disasters with a major agricultural component, such as fires, floods, landslides and tsunamis8.

8.2.4.4Earthquakes and tsunamis

8.2.4.3Local effects


________

8 Not all tsunamis are caused by earthquakes on land. The tsunami which badly hit 120 km2 of

coastal areas in West Sepik (Papua New Guinea) on 17 July 1998 was due to an earthquake that

occurred in the sea.

The 1995 (17 January) earthquake at Kobe triggered widespread fires whichcontributed significantly to the estimated US$ 100 billion damage. About 107 m3

of debris were produced, which had to be disposed of. Also in Japan, the 1933(2 March) earthquake at Miyagi, killed the same number of people (3 000) as aresult of the tsunami that was caused by the earthquake.

Regarding direct losses, an interesting set of data is provided by Rantucci (1994)regarding the earthquake that took place on Luzon on 16 July 1990 (magnitude 7.7on the Richter scale). Unusually large slip motions (up to 6.2 m in amplitude) wererecorded, but most damage was due to liquefaction9 and landslides. The earthquakeis seen as part of the increased geological activity in the area which also includesthe Pinatubo eruption.

Table 8.1 is interesting in that it compares the impact of the earthquake withthe eruption of Mount Pinatubo which occurred in virtually the same area and atthe same time of the year. Thus, the agricultural context is very comparable for bothevents.

Only 8.9 per cent of the total damage occurred in the agricultural sector,distributed as follows: crops, 38.6 per cent; fisheries, 28.6 per cent and irrigation, 7 per cent.

Rantucci also provides data on other disasters which have affected thePhilippines10. During the period 1987 to 1991, the damage from cyclones amountedto US$ 1.5 billion, the same order of magnitude as that caused by the earthquakeand the Pinatubo eruption of 1990 and 1991.

Snowstorms and avalanches normally occur outside the cropping season and, therefore, do not normally affect crops, except winter crops. However people,cattle, pasture, etc. are at risk. The February and March 1995 snowstorm in Nagquprefecture (northern Tibet) affected 130 000 people, although financial damagewas relatively limited due to the low development level of the region. Grasslandswere hit badly by what was estimated to be the worst snowstorm in 50 years, leaving hundreds of people and just under 3 million head of livestock stranded andin danger of freezing to death.

One of the worst recorded avalanches occurred in 1970 in Peru, where 18 000people died.

8.2.4.5Snowstorms and avalanches


Table 8.1Losses (million US$) due to the July

1990 earthquake and the eruptionof Mount Pinatubo in the Philippines.

Data from Rantucci, 1994

July 1990 Pinatubo eruptionearthquake June 1991

M US$ % M US$ %Crops 22.0 38.6 44.7 10.5Fisheries 16.3 28.6 2.3 0.5Livestock/poultry 1.6 2.8 4.8 1.1Irrigation 4.0 7.0 10.6 2.5Forestry 177.9 41.9Others, incl. infrastructure 13.1 23.0 4.6 1.2Foregone revenue 179.6 42.3

Agriculture total 57.0 8.9 424.5 59.7

Infrastructure 273.8 43.0 66.5 9.3Private property 158.2 24.8 205.1 28.8Industry/commerce 104.0 16.3 15.3 2.2Mining 21.1 3.3 0.0Tourism 22.9 3.7 0.0

Total 637.0 100.0 711.4 100.0

________

9 Some earthquakes cause fine-grained materials such as sand to behave like liquids, causing

buildings and other objects to sink into the ground.

10 The cost to the Philippines (returning residents, increased oil price) of the 1990–1991 Gulf Crisis

amounts to 380 million US$. The GNP of the Philippines was 28.6 billion US$ in 1990.

Landslides and mudslides are usually associated with heavy rains, but also withearthquakes, dam failures and volcanic eruptions. In the latter case, it is often rainfall on recently fallen volcanic ash that constitute the main source of the disaster, often entombing whole villages and sweeping away crops and agriculturalland.

Two Central American examples of land- and mudslides associated withearthquakes are the following: in 1986 (10 October) in San Salvador, an earthquakeand landslide made 300 000 homeless; in 1987 (5 and 6 March) an earthquake innorth-east Ecuador was followed by mudslides which buried several villages.

One of the most well known examples of mudslides associated with a volcaniceruption occurred in 1985 (13 November) in Armero, Colombia, after the eruptionof the Nevado del Ruiz, when 25,000 people died. The village of Armero was totallyengulfed by a torrent of mud when the La Lagunilla River burst its banks. Around11 000 ha of agricultural land was ruined. Snowmelt due to the lava flow was one ofthe main causes of the disaster (Nardin, 1989).

Landslides typically occur after long spells of heavy rain in areas of hilly terrainand marked seasonal rainfall patterns (dry/wet), as in Uttar Pradesh (India) duringAugust 1998. At this time 200 people perished and “terrace cultivated” crops andthe terraces themselves underwent serious damage.

Databases of disasters include a number of examples of dam failures, and it is surprising that the subject does not receive more attention. In many cases, dams arenot made from concrete and simply take the form of storage reservoirs holding lessthan 5 Gm3 of water which is used for irrigation and drinking water. The failure ofthese structures leads to loss of land and often to loss of habitat, and sharply reducesirrigation potential.

Some of major dam failures have occurred in China, for instance in 1975(August) at Banqiao and Shimantan, Huai River, Henan province. The number ofdeaths was 230 000. The two dams had been built in the 1950s. Within two hours ofthe nearly simultaneous bursts 85 000 people had died. An additional 145 000succumbed in ensuing epidemics and famines.

In 1993 (27 August) there was a dam failure in Qinghai province, westernChina, north of Tibet. Although it held only 2.6 Gm3 of water, the breaching of thedam at the Gouhou reservoir caused “big losses in lives and property” to nomadicherders and farmers in the semi-arid high-elevation plateau region.

Epidemics are mentioned because they are frequently triggered by unfavourableconditions at least partially due to weather. The transmission of diseases – especially vector-borne diseases – is weather dependent. In addition epidemics areassociated with other disasters such as floods. An example was mentioned aboveunder the 1993 Gouhou dam failure.

Torrential storms in the early 1300s in Asia are often quoted as one of theremote causes of the plague which began in China and eventually reached Europe inthe 1340s via trade routes. The disastrous black plague resulted which killed onethird of the European population (McNeill, 1989).

Human epidemics are mentioned also because they interfere with almost allfarming activities, frequently giving rise to secondary famine. For a general overviewof human health and climate, refer to McMichael, et al., 1996.

Although they may affect agriculture, often with an indirect weather component,the following are probably not relevant in quantitative terms in the current context: transportation accidents (aeroplane crashes, railroad accidents), miningaccidents and building collapse in urban areas.

Several cases of industrial explosions are known to have severely affectedpopulations and farming. For instance, the 1984 (2 December) methyl isocyanategas leak from a pesticide plant in Bhopal (India) killed 3 000, injured 100 000 andaffected about 250 000 people. Agricultural impacts were limited (7 000 cattlekilled) but damage to the natural environment remains largely unassessed(Shrivastava, 1996).

8.5.6VERY RARE FACTORS

8.5.2OTHER NON-GEOPHYSICAL FACTORS

8.2.4.7Dam failtures


The well documented dioxin release at Seveso (Italy) on 1976 (10 July) affected36 000 people but killed large numbers of domestic fowl and pigs population. Themost affected area has been sealed off ever since (posing some novel institutionaland legal problems). It is estimated that the dioxin will have biodegraded by 2040(De Marchi, 1996).

A very rare event occurred in June 1908 at Tunguska where a meteorite impacttook place in a deserted area of Siberia. The explosion of a stony asteroid, probably10 km across, took place 10 km above the ground (roughly equivalent to 15 megatonnes of explosive, or 1 000 Hiroshima bombs), levelling 2 000 km2 of forest; itsradiance caused widespread fires.

8.3 INFORMATION TO BE STORED IN THE DATABASE

The following is a proposal for a database of extreme agrometeorological events. Asindicated in the introduction, we adopt a somewhat broader approach – that of adatabase of agricultural disasters of geophysical origin, including man-made disasters with an atmospheric or hydrological component.

The basic idea behind this section is that a database of extremeagrometeorological events should comprise three separate, but cross-linked, sectionsfor each “event”:

(a) The precise description of the geophysical factors that caused the event;(b) The precise description of the impacted system before the event; and(c) The precise description of the impact (losses), in quantitative terms.

All variables will be geo-referenced.

The “event” is the elementary database unit (record). A disaster can be regarded as an event, or as a succession of events. For instance, the 16 July 1990 Luzon earthquake can be treated as a whole, or each of the successive shocks, which lastedfor months, can be treated as a separate event.

The event thus defines the timescale of the database unit. In addition, thespatial scale must be provided: a disaster can be analyzed for the sake of convenienceby administrative units or by physiographic or other logical units, including agro-economic or agro-ecological zones.

The event must be “impact oriented” to avoid difficulties when the source andthe target of complex events are different. For instance, the “Chernobyl accident”occurred at precise geographic coordinates (130 km north of Kiev), but the impact,which constitutes the raison d’être of the database occurred all over Europe.Therefore, the event would be best defined a “Chernobyl impact in Ukraine”,“Chernobyl impact in Germany”, etc.

It might also be decided to treat impacts and extreme factors separately,particularly if impacts of the same extreme factors have been felt at different timesand in different areas. This will, however, require a very comprehensive spatio-temporal description of the extreme event (which is not always possible or relevant).

This is the conventional name of the event, for instance “1845–46 Irish potatofamine”, or “Chernobyl nuclear accident”.

The typology of extreme events has been discussed. It constitutes an essentialdescription of the event. Included in this description must be references to thecomplexity of the event and the time and spatial scales.

This can be a point or a polygon of coordinates.

The timing and duration of the event must be given including any long-term effectswhich sometimes extend over decades.

8.3.1.4Timing of event

8.3.1.3Location of event

8.3.1.2Type of event

8.3.1.1Name/code of event

8.3.1DEFINITION OF AN EVENT


This item provides the opportunity of vertical and horizontal links. Vertical linkscan point to a major causal event (e.g. “1997–98 El Niño”), while horizontal linkswould include pointers to events due to the same cause, for instance tropicalcyclone Gilbert (9–19 September 1988) affected Central America, Jamaica, GrandCayman, Yucatan, Mexico, Texas and a number of other locations – which may allhave been treated as separate events. Or the tropical cyclone which affectedBangladesh on 24 and 25 May 1985 may have been dealt with as a separate eventfrom the accompanying tsunami, which would make a lot of sense since the affected areas were both different and rather well circumscribed.

The wording “thematic” refers to the agricultural or environmental description ofthe impacted system, the spatial extent of which was covered under 8.3.1.3. Thisitem constitutes one of the weak links in current impact assessments.

In a previous CAgM report, we underlined the factors which must be taken intoaccount (Gommes, 1997), in particular the types of crops (varietal information ifpossible) and the stage of growth.

It is essential that long-term effects, including recovery, be adequately covered.The role of agricultural research stations in extreme-factor prone areas cannot

be over-emphasized as they constitute one of the most valuable sources of data forquantitative impact data, risk assessments and impact forecasting. Contrary to acommon practice, observations must be continued after the event hits the station.This is because 100 per cent loss is very rare. Many crops somehow recover undernew agronomic and phytosanitary conditions; a situation which is difficult to modelin the absence of data.

As with the description of the production system (8.3.2.1), it will be difficult to provide specific items under this heading until such time as an agreed typology isavailable. It should be stressed that the dynamic aspects of extreme factors must beproperly covered, including an estimate of return periods.

In the absence of actual measures of the intensity of the most violent extremefactors such as tropical cyclones and earthquakes, impact-oriented intensity indicesconstitute invaluable tools. These indices can combine several aspects of thedestructive power of the factor or a complex of factors. Palmer’s Drought Index, orthe Sapphir-Simpson Scale used for hurricanes are more useful in quantitativeimpact and risk assessment than for pure geophysical measurement. It is oftenoverlooked that many common “scales” (Mercalli and Richter scales forearthquakes, or the Beaufort wind scale) are empirical impact-oriented scales.

The less violent extreme factors (drought, cold spells) can usually be expressedin terms of normal routine meteorological observations, although the above-mentioned indices are often very relevant as well.

Impact assessments currently constitute one of the weakest points on the path leading to a more quantitative and systematic approach. There does not appear tobe a standard methodology and, even when assessments have been carried outobjectively, the description of the impacted system and of the extreme factor areoften insufficient. We stress again, the role of agricultural research stations in theacquisition of relevant data.

This includes both the immediate loss and the long-term losses of agricultural production.

Damage indices would be extremely useful, for instance the Typhoon DamageIndex for crops (DIc) quoted by Jose (1994):

DIc = 0.37 V1.11

where V is the sustained windspeed in km/h. Similar indices are available for anumber of applications, including ozone effect on wheat (Finnan, et al., 1997).

Production loss

8.3.2.3Third component: impact

assessment

8.3.2.2Second component: description of

extreme factor

8.3.2THE THREE DATABASE COMPONENTS

– THEMATIC DESCRIPTION OF

IMPACTED SYSTEM

8.3.2.1First component: description of the

production system

8.3.1.5Links to other events


We include soil erosion, loss of biodiversity, etc. under this item.

Essentially the financial loss by agricultural sectors.

Other aspects are important. It would be useful to include an analysis of how thedisaster was managed, how the management increased or reduced losses, the quality of assessments that were done at the time, how the government managedthe crisis – suppression, confidentiality, recognition, amplification, etc.

Understanding the sources of data is crucial. Comprehensive impact assessments arerarely available. Much contradictory information becomes available via the mediawhile the event is happening but final figures are rarely broadcast. Impact assessments prepared by governments are often biased because they are conductedimmediately after events in order to obtain assistance. Among other shortcomings,they rarely include long-term effects. It is common to find that different sourcesquote vastly different estimates of casualties, particularly for events that have affected extensive areas and more than one country (a good example is the tropicalcyclone that affected West Bengal and Bangladesh on 12 and 13 October 1970,where the total number of casualties is virtually impossible to determine).

Part of the problem stems from the lack of commonly agreed terminology andtypologies. Take “people affected”, an indicator occurring in most existing disasterdatabases and, in one form or another, in most descriptions and analyses of disasters.The same applies to “hectarage affected” or “agricultural production loss” – not onlydo different authors adopt different definitions, but the area, to which the data refer,usually corresponds to administrative units where an official emergency has beendeclared, ignoring adjacent districts.

Finally, extreme care must be exercised for many man-made disasters, wherenon-technical motivations often dominate and for which a critical assessment ofthe reliability and neutrality of the sources is essential.

8.4 CONCLUSIONS AND RECOMMENDATIONS

As indicated in the introduction, it is suggested that a database of extreme agrometeorological events does not make much sense, as the event becomes“agrometeorological” only when it affects agriculture, i.e. when we have a disaster.The database should thus be one of agricultural disasters resulting from extremegeophysical and man-made factors with an atmospheric component.

Almost all extreme events are likely to affect agriculture, although geophysicalfactors, by their sometimes large geographic extent, have the potential to lead tothe greatest damage in terms of instantaneous and medium-term production losses.

Regarding methodology, the following points are underlined:(a) The need to develop a proper typology of impacts as the first step in the definition

of data requirements and the improvement of impact and risk assessments and forecasting.

(b) The need to develop two types of indices, by typology, as tools for impact and riskassessments. One index describing the “global” intensity of extreme factors in thewithin a specific category of disaster; and the other relating elementary extremeevents (for example maximum instantaneous wind speed) with observed agricultural damage.

(c) The need for agricultural research stations and agrometeorological stations to continue and intensify their observations after the occurrence of extreme events,in order to provide badly needed quantitative impact and factor data sets.

(d) Regarding the structure of the database, there should be three separate, but cross-linked, building blocks:(i) The precise description of the geophysical factors that caused the event;(ii) The precise description of the impacted system before the event; and(iii) The precise description of the impact (losses) in quantitative terms.All variables should be geo-referenced.

8.3.3SOURCES OF DATA

Other aspects

Socio-economic impact

Environmental losses


• Ayers, R.S., and Westcot, D.W., 1976: Water quality in agriculture. FAOIrrigation and Drainage Bulletin No. 29, FAO, Rome, 81 pp.

• Besoain, M.E., Spulveda, W.G. and Sadzawka, R.A., 1992: La erupcion delvolcan Lonquimay y sus efectos en la agricoltura. Agricultura Technica(Santiago), 52(4):354–358.

• Bourke, A., and Lamb, H., 1993: The spread of potato blight in Europe in 1845–6and the accompanying wind and weather patterns. Irish Meteorological Service,Dublin, 66 pp.

• Briffa, K.R., Jones, D.D., Schweingruber, F.H. and T.J. Osborn, 1998: Influenceof volcanic eruptions on northern hemisphere summer temperature over thepast 600 years. Nature, 393:450–455.

• Chester, D.K., Duncan, A.M., Guest, J.E. and Kilburn, C.R.J., 1985: MountEtna, the anatomy of a volcano. Chapman and Hall, London. 404 pp.

• Davis, N.Y., 1996: The Exxon Valdez oil spill, Alaska. In: The long road to recovery. Mitchell, J.K. (ed.), UNU Press, Tokyo, pp. 231–272.

• De Marchi, B., Funtowicz, S. and Ravetz, J., 1996: Seveso: a paradoxical classicdisaster. In: The long road to recovery. Mitchell, J.K. (ed.), UNU Press, Tokyo,pp. 86–120.

• De Silva, S.L. and Zielinski, G.A., 1998: Global influence of the AD 1600 eruption of Huaynaputina, Peru. Nature, 393:455–458.

• ECE, 1997: Effects of long-range transboundary air pollution. ECE Air pollutionstudies No. 13, ECE, Geneva, 53 pp.

• Finnan, J.M., Burke, J.I. and Jones, M.B., 1997: An evaluation of indices thatdescribe the impact of ozone on the yield of spring wheat (Triticum aestivum L.).Atmospheric Envir., 31(17):2685–2693.

• Gommes, R., 1997: An overview of extreme agrometeorological events. In:Benson, G.J., Dambe, D., Darnhofer, T., Gommes, R., Mwongela, G.N.,Pedgley, D.E. and Perarnaud, V., Extreme agrometeorological events. CAgMReport No. 73, TD No. 836, WMO, Geneva, pp. 1–9.

• Gommes, R., 1998: Climate-related risk in agriculture. In: Expert meeting on riskmanagement methods. Toronto, 30 April–1 May 1998, IPCC, pp. B1–B13.

• Hansen, J., et al. 1996: A Pinatubo climate modeling investigation. In: (Fiocco,G., Fua, D. and Visconti, G. (ed.), The Mount Pinatubo eruption: effects on theatmosphere and climate. NATO ASI Series Vol. I 42, Springer-Verlag.Heidelberg, Germany, pp. 233–272.

• IFMR, 1994: Sharing the challenge: floodplain management into the 21st century.Report of the Interagency Floodplain Management Review Committee to theAdminstration Floodplain Management Task Force, Washington.

• Jose, A.M., 1994: Climatological information on tropical cyclones in thePhilippines. In: Report of the Conference on Tropical Urban Climates. 28 March–2 April, Dhaka, Bangladesh. WCASP No. 30, TD No. 647, WMO, Geneva, pp.149–160.

• Kersteins, K., Federholzner, R. and Lendzian, K.J., 1992: Dry deposition andcuticular uptake of pollutant gases. Agric., Ecosystems and Envir., 42:239–253.

• Larson, N.R., Michalsky, J.J. and LeBaron, B.A., 1996: Rattlesnake MountainObservatory (46.4 N, 119.6 W) multispectral optical depth measurements:1979–94. CDIAC Spring 1996 Bulletin, p. 15.

• Marples, D.R., 1996: The Chernobyl disaster: its effects on Belarus andUkraine. In: The long road to recovery. Mitchell, J.K. (ed.), UNU Press, Tokyo,pp. 183–230.

• Marshall, F., Ashmore, M. and Hinchcliffe, F., 1997: A hidden threat to food production: air pollution and agriculture in the developing world. Gatekeeper seriessustainable agriculture programme, International Institute for Environment andDevelopment No. 73, 24 pp.

• McCormick, P., Tomason, L.W. and Trepte, C. R., 1995: Atmospheric effectsof the Mount Pinatubo eruption. Nature, 373:399–404.

• McGuire, W.J., 1997: Volcanic disasters. Past, present, future. Science Progress,80(1):83–99.

REFERENCES


• McMichael, A.J., Haines, A., Sloof, R. and Kovats, S. (eds.), 1996: Climatechange and human health. WHO, Geneva, 297 pp.

• McNeill, W., 1989: Plagues and people. Anchor Books, New York, 340 pp.• Mitchell, J.K. (ed.), 1996: The long road to recovery. UNU Press, Tokyo, 307 pp.• Nardin, D., 1989: Una catastrofe annunciata: l’alluvione provocata dal vulcano

Arenas nel Nevado del Ruiz-Colombia (13 Novembre 1983). Annali Acc. Ital.Sci. Forestali, 38:459–475.

• Rantucci, G., 1994: Geological disasters in the Philippines, the July 1990 earthquakeand the June 1991 eruption of Mount Pinatubo. Italian Ministry of ForeignAffairs, Directorate General for Development Cooperation, Rome, Italy, 154 pp.

• Sagan, C. and Turco, R., 1991: L’hiver nucléaire. Seuil, Paris, 434 pp.• Schönwiese, C.D., 1988: Volcanic activity parameters and volcanism-climate

relationships within recent centuries. Atmosfera (Mexico), 1(3): 141–156.• Shrivastava, P., 1996: Long-term recovery from the Bhopal crisis. In: The long

road to recovery. Mitchell, J.K. (ed.), UNU Press, Tokyo, pp. 121–46.• Smith, R., 1992: Catastrophes and disasters. Chambers, Edinburgh, 246 pp.• Stiegeler, S.E., 1976: A dictionary of earth sciences. Macmillan Press, London,

301 pp.• Stommel. H. and Stommel, E., 1979: The year without a summer. Sci. Am.,

240(6):134–140.• Susman, P., Keefe, O. and Wisner, B., 1983: Global disasters, a radical

interpretation. In: Hewitt, K., (ed.), Interpretations of calamity, The Risks andHazards Series: 1. Allen & Unwin Inc. Boston, pp. 263–280.

• Wirth, E., van Egmond, N.D. and Suess, M.J. (eds), 1987: Assessment of radiation dose commitment in Europe due to the Chernobyl accident. Published onbehalf of WHO, Regional Office for Europe, by Institut für Strahlenhygiene(ISH) des Bundesgesundheitsamtes, Munich, ISH N. 108. 51 pp.

• Youxue Zhang, 1996: Dynamic of CO2-driven lake eruptions. Nature,379:57–59.


CHAPTER 9

CONCLUSIONS AND RECOMMENDATIONS(by H.P. Das)

9.1 CONCLUSIONS

Extreme events due to both atmospheric and non-atmospheric factors take a heavytoll in deaths and inflict much suffering, and there are consequent substantial lossesin productivity, including to crops and livestock. Extreme natural events illustratean important aspect of the complex process by which people interact with biological and physical systems. Every parameter of the biosphere is subject to seasonal, annual, or daily fluctuations – constituting hazards to people. But humanadjustments to the frequency, magnitude, or timing of extremes are based on ratherimperfect or inadequate knowledge. If there were perfectly accurate predictions ofwhat would occur and when it would occur in the intricate web of atmospheric,hydrologic and biological systems, there would be no hazard. However, there wouldstill remain the question of how to respond effectively to the completely predictable order of events. Ordinarily, extreme events can only be foreseen asprobabilities whose time of occurrence is unknown.

The hazard accompanying the occurrence of the rains for dry-land farmers orthe duration of peak river flow for a floodplain located manufacturer or themagnitude of the infrequent but certain earthquake for a fault-zone dweller is asignificant element in decisions which many individual users of the environmentmust make on a daily, seasonal, or yearly basis. The more common extremegeophysical events are avalanche (snow), coastal erosion, drought, earthquake,flood, frost, hail, landslide, lightning, snow, tornado, tropical cyclone, volcano andwind.

The study of agricultural meteorology or climatology has enabled people to havean insight into the effects of weather and climatic elements on agriculture. The roleof various extreme climatic events that affect agricultural production negatively hasnow been understood to some extent. In themselves, weather extremes are notnecessarily hazardous, although they may become so if they prevail for prolongedperiods of time. This accumulative effect of weather extremes is evident in cases ofdroughts, heatwaves and floods. The atmospheric factor may fulfil a variety of rolesin the development of hazard situations. A broad distinction can be made betweenphenomena such as tropical cyclones, tornadoes and lightning which involve thesudden impact of massive amounts of energy discharged over an extremely shorttime, and those features which become hazards only if they exceed tolerablemagnitudes within or beyond certain time limits. In this latter category can beincluded wind hazards associated with extra tropical low pressure systems,heatwaves, snow, heavy rain, frosts and droughts.

Thresholds and qualitative effects characterize a number of plants and animalswith regard to their response to weather factors. Well-known examples are the effectof temperatures on rice sterility and the breaking of stems and branches of certainrubber cultivars by wind. Windstorms and tropical storms (hurricanes and typhoons)with very high winds can destroy fields of cereal within minutes and reduce theyield. It is also observed that plantations suffer more direct damage than naturalforests. Of course, root and tuber crops and creeping plants suffer very little fromhurricanes/tropical storms, while tree crops and cereals may be badly hit. Similarlyfloating rice varieties are characterized by very fast growing stem elongations whichcan keep pace with rapidly rising water during floods.

Other hazardous climatic events include widespread and local thunderstorms,tornadoes, squall lines, hailstorms and weather related wildland fires whichsometimes cause havoc to agriculture and forestry. Climate and its extremes are alsoresponsible for the growth and development of pests and locusts which can badly

affect plants. Earthquakes, volcanic eruptions and landslides, although belonging togeological events, are likely to result in losses to forestry or agricultural operations.

By taking anti-disaster preparedness measures in advance and thereafter byundertaking systematic post-disaster rehabilitation, the effects of most of thedreadful natural calamities can be mitigated considerably. Valuable work in thisdirection is already being done in many countries. Although natural calamitiescannot be averted, their destructive impact in terms of loss of human and animallife and the upset of the ecological balance can, no doubt, be considerablyminimized. Planning and management for the prevention and mitigation of extremeevents are matters of vital significance for the safety and well-being of millions ofpeople who inhabit the globe in disaster prone areas. In addition to local andnational action, international and regional cooperation should be promoted forenhanced prevention and mitigation.

On account of the vast similarities of situations arising out of geographicalfactors, social and economic conditions, it would be advantageous to undertakespecific case studies of work done in the prevention and mitigation of extremeevents in different countries of the Asian and Pacific regions in particular, whichare more prone to frequent disasters than to other regions. This will bring outfeatures common to a number of countries, which can be duly considered.

Finding out how these responses to extreme events differ from place to placeand from time to time helps understand the way one system affects another. It alsoalllows us to know how these relationships can be changed for the benefit of thepeople who suffer from severe events. If the means of enabling individuals to takeintelligent action or governments to design and carry out effective programmes areto be improved, it is essential to gain greater knowledge of the processes by whichpeople do, in fact, cope with the hazards of nature. That is the aim of thecollaborative programme of research on natural hazards.

9.2 RECOMMENDATIONS

(1) There is a need for the maintenance of observational networks and theirenhancement to accurately depict extreme events and their impact on agriculture.

(2) Efforts should be made to strengthen the links between the information generators and the users of agrometeorological information and provide training to information users.

(3) Agrometeorological services in developing countries should be strengthenedthrough the placement of more automatic weather stations to overcome theproblems of data quality and observer biases. Remotely sensed data can beused to fill the gaps, where they exist.

(4) Reliable ground networks are still necessary in the light of the need to provideground truth for remotely sensed observations.

(5) Radar coverage should be extended to more areas affected by extreme events.

(1) In addition to longer term strategies, short-term remedial programmes shouldbe put in place for dealing with immediate problems such as soil erosion, salinization or famine, designed to alleviate their more immediate manifestations.

(2) The monitoring of certain extreme events should be undertaken continuouslythroughout the year and over the entire country. In this way early warningscan be made of anomalies in the hydrological conditions of the soil, the spatial extent of such anomalies can be plotted and governments can be alerted.

(3) In the case of extreme events such as tropical cyclones, warnings should beregularly updated and be more specific regarding time and place of landfall,maximum wind strength to be expected, the expected intensity of rains andthe areas most liable to storm surge. By arrangement with the authorities,warnings should be so worded as to indicate clearly the nature of the actionthat should be considered by those to whom the warning is issued.

9.2.2MONITORING, EARLY WARNING AND

REMEDIAL MEASURES

9.2.1INFORMATION SYSTEMS


(4) In the event of disasters such as earthquakes, volcanic eruptions, floods,cyclonic storms, etc. it would be appropriate for each local authority to set upa permanent “Disaster Preparedness Committee” which would include appropriate experts from every sphere of life.

(1) Standard methods of impact assessment should be developed and disseminatedfor use by meteorological services.

(2) Practical applications should be generated from successful case studies on waysto combat extreme events, taking into full account matters of environmentand sustainable development.

(1) Given emerging developments in the monitoring of extreme events, the evaluation of their impacts and the need for a rapid diffusion of this information, training sessions should be organized to train trainers, development agencies and NGOs and end users.

(2) There is a need to raise environmental awareness among the local population.Local environmental issues should be integrated into school curricula andlocal methods should be employed in the monitoring of the management ofextreme events.

(1) Collaboration with CIMO, FAO, GCOS and with conventions such asUNCCD and UNFCC should be enhanced.

9.2.5COLLABORATION AND

COOPERATION

9.2.4TRAINING, EDUCATION AND

INCREASED AWARENESS FOR THE

GENERAL PUBLIC AND DECISION

MAKERS

9.2.3METHODOLOGY DEVELOPMENT

CHAPTER 9 — CONCLUSIONS AND RECOMMENDATIONS 137

Recent WMO Technical Notes

No. 160 Soya bean and weather. By F. S. da Mota.

No. 166 Quantitative meteorological data from satellites. By C. M. Hayden, L. F. Hubert, E. P. McClain andR. S. Seaman. Edited by J. S. Winston.

No. 172 Meteorological aspects of the utilization of solar radiation as an energy source.

No. 173 Weather and airborne organisms, By D. E. Pedgley.

No. 180 Weather-based mathematical models for estimating development and ripening of crops. By G. W. Robertson.

No. 181 Use of radar in meteorology. By G. A. Clift, CIMO Rapporteur on Meteorological Radars.

No. 182 The analysis of data collected from international experiments on lucerne. Report of the CAgM Working Groupon International Experiments for the Acquisition of Lucerne/Weather Data.

No. 184 Land use and agrosystem management under severe climatic conditions.

No. 185 Meteorological observations using navaid methods.

No. 186 Land management in arid and semi-arid areas.

No. 187 Guidance material for the calculation of climatic parameters used for building purposes.

No. 188 Applications of meteorology to atmospheric pollution problems. By D. J. Szepesi, CCl Rapporteur onAtmospheric Pollution.

No. 189 The contribution of satellite data and services to WMO programmes in the next decade.

No. 190 Weather, climate and animal performance. By J. R. Starr.

No. 192 Agrometeorological aspects of operational crop protection.

No. 193 Agroclimatology of the sugar-cane crop. By B. C. Biswas.

No. 194 Measurements of temperature and humidity. By R. G. Wylie and T. Lalas.

No. 195 Methods of interpreting numerical weather prediction output for aeronautical meteorology. Report of the CAeMWorking Group on Advanced Techniques Applied to Aeronautical Meteorology.

No. 196 Climate variability, agriculture and forestry. Report of the CAgM-IX Working Group on the study of ClimateEffects on Agriculture including Forecasts, and of the Effects of Agriculture and Forests on Climate.

No. 197 Agrometeorology of grass and grasslands for middle latitudes. By A. J. Brereton, S. A. Danielov and D. Scott.

No. 198 The effect of temperature on the citrus crop. By Z. Gat, Y. Erner and E. E. Goldschmidt.

No. 199 Climate variability, agriculture and forestry: an update. By M. J. Salinger, R. Desjardins, M. B. Jones,M. V. K. Sivakumar, N. D. Strommen, S. Veerasamy and Wu Lianhai, CAgM Rapporteurs on the Effects ofClimate Change and Variability on Agriculture and Forestry.

No. 200 Climate variability, agriculture and forestry: towards sustainability. By M. J. Salinger, R. L. Desjardins, P. H.Karing, S. Veerasamy and G. Zipoli, (CAgM-XI Joint Rapporteurs on the Effects of Climate Change andVariability on Agriculture and Forestry – Agrometeorological Aspects of Management Strategies andImprovement of Sustainability).